EPA-650/2-74-007
Environmental Protection Technology Series
January 1974
m$
sgf^p^pxiii?;;pi^i^^^^s;g
iipxiiiiSSSiv'iv^^
,..,... .'.'I
I
-------
EPA-650/2-74-007
PARTICULATE EMISSIONS
FROM ALFALFA DEHYDRATING PLANTS
CONTROL COSTS AND EFFECTIVENESS
by
Kenneth D. Smith
American Dehydrators Association
5800 Foxridge Drive
Mission, Kansas 66202
Grant No. R801446
ROAP No. 21ADJ-83
Program Element No. 1AB012
EPA Project Officer: G.S. Haselberger
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
January 1974
-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
ii
-------
ABSTRACT
This report presents the results of an extensive field-testing program
to characterize particulate emissions from alfalfa dehydrating plants and
to evaluate the cost/effectiveness of available control methods.
Testing was conducted during the growing seasons of 1971, 1972, and 1973
at fourteen plants in Kansas, Nebraska and Colorado.
In the first testing phase (1971), cyclone effluents were sampled to
determine mass rates and size distributions of particulates emitted
from the drying, grinding, and pelleting operations, under measured
process operating conditions. Emission factors are presented for each
of the unit operations. Test results show that emissions from the
drying operation comprise more than 75% of the total emissions, and
are the nost difficult to control. Dryer emissions vary with process-
weight-rate, hay quality, dryer operation, and cyclone collector efficiency.
Typically, a 60% reduction in dryer emissions is required to meet the
common process-weight-rate standard.
In the second testing phase (1972), benchmark performance data were
obtained on two pilot-scale and three full-scale wet scrubbers and on
two full-scale control systems which recycle effluent from the primary
cyclone. During each performance test, process operating conditions
were monitored. Performance data for the wet scrubbers include fractional
collection efficiency, gas throughput, pressure drop, and water usage.
Test results indicate that medium efficiency wet scrubbers have the poten-
tial to bring alfalfa dryer emissions into compliance with process-weight-
rate standards, although problems of water clarification and sludge
disposal remain to be solved. The results also indicate that the partial
recycle of primary cyclone effluent back to the dryer furnace holds promise
for the significant reduction of particulate emissions, and may provide a
substantial fuel savings.
In the third phase (1973), full scale control devices/systems were tested
to determine the effectiveness and freedom from operating problems of an
intermediate pressure drop scrubber and recycle systems, and the effective-
ness of plant modifications and operating procedures in reducing
particulate emissions. Periodic source testing was performed under
monitored process conditions although testing was limited to particulate
rate measurements at the outlets of control devices/systems. In the case
of wet collectors, equipment and methods for water treatment and sludge
disposal were also studied. The results of this phase demonstrate
that dehydrating plants can be operated in compliance of emission standards
through plant modifications and operating procedures. The results further
show that partial recycle of the primary effluent back to the furnace and/or
the use of wet scrubbers are very effective methods of substantially reducing
the particulate emissions. However, the results suggest that neither
recycle systems njr wet scrubbers have the capability of bringing a grossly
dirty plant into compliance, i.e., a plant which produces excessive smoke
as a result of poor operating procedures and/or process controls.
iii
-------
Based on cost figures submitted by the control equipment manufacturers,
equipment and installation costs for a wet scrubber and water treatment
system range from $35,000 to $45,000 and annual operating costs are
about $2,000. The Installed cost of a recycle system Is about $20,000
and a net operating savings of $2,000 is realized because of decreased
fuel usage. These figures apply to a model control problem which requires
a control efficiency of about 60% for an effluent of 30,000 acfm.
This study was funded by the U. S. Environmental Protection Agency and by
the American Dehydrators Association. Midwest Research Institute
conducted the field tests and performed part of the engineering and cost
evaluations.
iv
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 3
III Introduction 7
IV Objectives 11
V Description of Control Equipment 13
Fisher-Klosterman 13
APPCOR 13
Koch 17
Air Conditioning Corp. 20
Thompson 20
Hell 23
Abilene Plant Modifications 25
Lawrence Plant Modifications 25
Neodesha & Berthoud Plant Modifications 29
VI Measurement Techniques, Equipment and 31
Procedures
VII Measured Performance Data and Control 37
Effectiveness
Fisher-Klosterman 37
APPCOR 37
Koch 41
Air Conditioning Corp. 46
Particle Size Distributions 49
Fractional Collection Efficiency 49
Summary of Wet Scrubbers 51
-------
Section
VII (cont.)
VIII
IX
X
XI
XII
CONTENTS - Continued
Page
Water Clarification 52
Thompson 54
Heil 58
Summary of Recycle 62
Effect of Recycle on Product Quality 63
Abilene 63
Neodesha and Berthoud 66
Comparison With the Bay Area Standards 66
Visual Opacity 72
Effect of Process Conditions 73
Single-Plant Correlation 73
Multiplant Correlation 74
Cyclone Collection Efficiency 76
Emission Factors 77
Control Costs - Model Problem 79
Sampling Log, Performance and Process 83
Data, and Plant Equipment Specifications
Acknowledgements 109
References 111
vi
-------
FIGURES
Page
1 Fisher-Klosterman Wet Collector 14
2 Neodesha Plant Interfacing With Pilot-Scale IS
Wet Collectors
3 APPCOR Wet Collector 16
4 Koch Wet Collector 18
5 Plant Interfacing for Full-Scale Wet Collectors 19
6 Thompson Recycle System 21
7 lopeka Plant Interfacing With Thompson Recycle 22
System
8 Grand Island Plant Interfacing With Heil Recycle 23
System
9 Dundee Plant Interfacing 24
10 Abilene Plant Interfacing 26
11 Lawrence Plant Interfacing 27
12 Neodesha and Berthoud Plant Interfacing 29
13 Straightening Vanes 35
14 Testing Platform 35
15 Testing Platform Side View 36
16 Particle Size Distributions - F/K Wet Collector 39
17 Particle Size Distributions - APPCOR Wet Collector 40
18 Particle Size Distrlbutuions - Koch Wat Collector 43
(Oxford)
19 Particle Size Distributions - Koch Wet Collector 44
(Rosel)
20 Particle Size Distributions - ACC Wet Collector 48
21 Fractional Collection Efficiency Curves 50
22 Particle Size Distributions - Thompson Recycle 56
System
23 Particle Size Distributions - Heil Recycle System 60
-------
FIGURES - Continued
Page
24 Control Effectiveness versus Bay Area Emission 68
Standard
25 Control Effectiveness versus Bay Area Emission 69
Standard
26 Control Effectiveness versus Bay Area Emission 70
Standard
27 Control Effectiveness versus Bay Area Emission 71
Standard
28 Calculated Fine Particle Emissions (Smaller Dryers) 75
viii
-------
TABLES
No. Page
1 Testing Program - Equipment and Plants 10
2 Process Parameters Monitored During Testing 32
3 Fisher-Klosterman Performance Data 38
4 APPCOR Peiformance Data 38
5 KOCH (Oxford) Performance Data 42
6 KOCH (Rozel) Performance Data 42
7 KOCH (Oxford) 1973 Performance Data 45
8 Air Conditioning Corporation Performance Data 47
9 Thompson Performance Data - Topeka 55
10 Thompson Performance Data - St. Marys 57
11 Hell Performance Data - Grand Island 59
12 Heil Performance Data - Dundee 61
13 Effect of Exhaust Gas Recycling on Carotene 64
and Xanthophyll Stability
14 Abilene Performance Data 65
15 Neodesha and Berthoud Performance Data 67
16 Visual Opacity Readings 72
17 Comparative Efficiencies of Primary Cyclone 76
Collectors
18 Uncontrolled Emission Factors 78
19 Model Problem Data Form - Performance and Cost 80
Data - Model Control Problem
20 Model Problem Cost Data 82
21 Sampling Log 84
22 Performance and Process Data Calculation Factors 88
23 Fisher-Klosterman/Neodesha Performance and 89
Process Data Dryer Conditions
24 APPCOR/Neodesha Performance and Process Data 90
Dryer Conditions
-------
TABLES - Continued
No. Page
25 Koch/Oxford Performance and Process Data 91
Dryer Conditions
26 Koch/Rozel Performance and Process Data Dryer 92
Conditions
27 ACC/Lexington Performance and Process Data Dryer 94
Conditions
28 Thompson/Topeka Performance and Process Data 95
Dryer Conditions
29 Hell/Grand Island Performance and Process Data 98
Dryer Conditions
30 Abilene Performance and Process Data 100
31 Koch/Oxford Performance and Process Data (1973) 101
32 Koch/Lawrence Performance and Process Data 102
33 Thompson/St. Marys Performance and Process Data 103
34 Neodesha Performance and Process Data 104
35 Berthoud Performance and Process Data 105
36 Hell/Dundee Performance and Process Data 106
37 Dryer Specifications 107
38 Bay Area Process Weight Table 108
-------
3Ecr;o:,
CONCLUSIONS
The source testing program conducted by the American Dehydrators
Association during the past three summers has provided a substantial
amount of information on the characteristics of particulate emissions
from alfalfa dehydrating plants and the effectiveness of available
control methods and systems. It now appears possible for each
dehydrator to plan =?rd schedule a technically and economically sound
strategy for the control of particulate emissions to meet the Bay Area
Process Weight Rate Standard.
The test program has shown that the quantity and size distribution of
emissions from the drying operation are strongly dependent on the follow-
ing conditions: a) percent of production capacity (based on the dry
throughput corresponding to rate of evaporative capacity and specified
moisture in the green chops); b) hay quality (protein, moisture content,
insect damage, age, foreign material, etc.); and c) the percentage of
moisture in the dry chops. Since low hay quality and high production rate
normally occur together, high productivity emissions may exceed low
productivity emissions manyfold.
Particle size distribution measurements have shown that the primary
cyclone emissions (drier emissions) are comprised of mechanically created
particles and heat generated particles. The particles created me-
chanically are generally larger than 1 micron and are referred to herein
as dust. The smoke or heat created particles are generally in the submicron
range.
As a consequence of drying, especially overdrying which frequently occurs
with high productivity, a very fine particulate smoke is generated. Although
the smoke may constitute only a small mass emission, it scatters light very
effectively and may frequently result in noncompliance with visual opacity
regulations. Moreover, the smoke readily penetrates conventional medium
efficiency control devices making effective control practically impossible.
Contrcl of dryer generated particulate emissions is also complicated by
the large volume of moist carrier gases.
During the testing program, it was learned that the visual appearance of a
plume is not a reliable Indicator of compliance with a mass limit regula-
tion. If overdrying occurs at a plant where the primary collector is
relatively efficient, the opacity limit might be exceeded even though
total mass emissions are in compliance with the mass limit regulation.
On the other hand, good dryer operation at a plant with an inefficient
primary cyclone **zy result in nencomplLance with the mass limit regulation
even though the plume (after steam dissipation) is in compliance with
opacity regulations.
Emissions from the grinding and pelleting operations consist of moderately
coarse particles carried by a relatively dry air stream. Additional fine
particles are generated by these operations if the alfalfa has been over-
dryed. At some plants, the hammermill is the production bottleneck, and the
material is overdryed to increase throughput.
-------
The testing program has demonstrated the capability of bringing alfalfa
dehydrating plants into compliance through process modifications and
operating procedures. Compliance was based on production conditions at
or above rated capacity and with average quality (17% protein) hay. The
question still remains of whether it is possible to operate in compliance under
conditions of low hay quality and high production rate.
The performance testing of wet scrubbers and recycle systems for the control
of particulate emissions has demonstrated that both types of systems are
effective in reducing emissions. However, in all cases, the test plants were
well "tuned" prior to testing of a control system so that uncontrolled
emissions were probably well below the average for the industry. Dryer
controls were checked and adjusted and dryers were carefully operated so
that the generation of smoke was minimized. In some cases, plants were
operated below rated capacity during testing. In other words, the results
are indicative of relatively clean plants which are not
representative of the industry as a whole.
Although performance test results are encouraging, operational problems
remain to be solved. The wet scrubber systems have attendant problems of
water treatment and sludge disposal - more testing and development work is
needed to provide practical and reliable solutions. The systems which
recycle effluent from the primary collector require more careful plant
operation and condensation in the recycle line can be a problem.
Recycle systems offer attractive fuel savings, but this advantage may be
easily exaggerated. Recycle of primary cyclone effluent back to the front
end of the dryer increases dryer efficiency at lower evaporation rates;
but, as the evaporation rate is increased, the amount of recycle must be
reduced to maintain sufficient excess air and the dryer itself becomes
more efficient. As a result, if the dryer is operated near capacity, the
fuel savings may become insignificant.
Equipment costs for the wet collectors, based on data submitted by the
manufacturers, are in the range of $15,000 to $25,000 including about
$5,000 for a water treatment system. Installation costs average about
$15,000, and annual operating costs average about $2,000 including
manpower for equipment maintenance.
Equipment and installation costs for retrofit of the recycle systems to an
existing plant are about $20,000. There may be a net savings in annual
operating costs because of decreased fuel usage.
The test data indicate that partial recycling of the primary cyclone
effluent back to the furnace had no effect on carotene or xanthophyll
stability under the conditions of these tests.
The total particulate emissions from an alfalfa dehydrating plant, without
emission control, average less than 20 Ibs./ton of pellet or meal produc-
tion.
-------
SECTION II
RECOMMENDATIONS
There are several options available for the control of particulate
emissions from alfalfa dehydrating plants. The first, most logical
approach, regardless of subsequent measures, is the proper maintenance
and operation of the alfalfa harvesting and dehydrating equipment.
Proper equipment operation is essential for the reduction of smoke
which once generated is virtually impossible to collect. Harvester
knives should be sharp and properly adjusted. Feeder discharge should
be uniform. Careful attention should be given to temperature controls
and balancing of gas flows. Furnace design and deficiencies should
be corrected where long flames and flame impingement are troublesome.
Cyclone collectors should be designed and operated to achieve acceptable
collection efficiencies. Hammermill capacity should be adequate and
hammer and screen maintained properly. These steps, plus some reduction
in plant throughput, especially under conditions of low hay quality,
have been demonstrated to bring plants into compliance. However, the
reduction of emissions by reducing throughput might well be the most
costly approach.
Partial recycle of the primary cyclone effluent back to the furnace
appears to be the second most logical step to controlling the
particulate emissions from alfalfa dehydrating plants. The test results
show that recycle can substantially reduce particulate emissions, may
reduce fuel consumption, and would reduce the net amount of gas to be treated
if a scrubber, or other control device, has to be added. There is a
potential for operating problems with recycle as a result of condensation in
the recycle lines. Recycle, and other operating procedures designed to
produce higher moisture dry chops (10% to 14%), may cause plugging of the
hammermill and other parts of the system unless the total plant is scaled
for this type of production.
The final step, if necessary, to bring a plant into compliance, would be to
install a medium energy wet scrubber. The test results indicate that a
scrubber with 4" to 6" of water pressure drop will bring the particulate
emissions down by at least 50%. If the particulate emissions consist pri-
marily of dust particles, a medium energy wet scrubber would almost
certainly bring the plant into compliance. Although the scrubber would
not be highly effective on the submicron smoke particles, it would do a
good job on the dust emissions and might improve the appearance of the
smoky plume considerably. There is a strong possibility that a plant
could not be brought into compliance with visual opacity regulations by
use of only a wet scrubber. Therefore wet scrubbers are not recommended
as the first control approach.
The above comments have dealt only with the control of emissions from the
primary cyclone. There is still the problem of particulate emissions from the
hammermill system and pellet cooling system. There should be no problem in
cleaning up the pellet cooler emissions through good engineering and well
designed equipment. In general, the best approach to controlling the
particulate emissions from the hammermill system cyclone is by direct
filtration with a bag house.
-------
The potential for fires in the bag house can, at least, be reduced with
the use of a "spark out" loop; i.e., extend the length of the air line
from the cyclone collector to the bag house sufficiently for an ignited
dust particle to have time to burn out before it reaches the bag house.
A 60* air line has worked satisfactorily, but may not be the optimum
length.
Hammermill emissions have been satisfactorily controlled by recycling a
large portion of the effluent from the hammermill system cyclone collector
back to the hammermill. The remaining portion of the air stream (15% to
30%) is usually bled off to the primary cyclone in order to prevent
excessive heat build up in the hammermill. This approach requires a good
degree of expertise if the hammermill recycle system is to function
properly and not cause operating problems.
Another approach which has been used in the industry to control hammermill
emissions is to discharge the hammermill cyclone effluent into the primary
cyclone. This technique has not been demonstrated through the testing
program to be an effective way of controlling the hammermill emissions
per se.
The following is a suggested check list for an alfalfa dehydrating plant
control strategy.
1. Survey and characterize plant; thoroughly assess the operating
condition of the plant including the performance characteristics
of all equipment, design compatibility, and maintenance require-
ments. Particularly important are: the drier-fuel control
system; the performance of the primary collector; and the
operational limit for dry chops moisture. Review plans and set
priorities for replacing equipment and generally upgrading the plant.
2. Upgrade plant operations; maintain and upgrade plant to achieve
good operating standards as determined in step 1. This is
particularly important with regard to drier controls and primary
cyclone collector performance. It may be necessary to increase
hammermill capacity to permit processing of dry chops having a
moisture content of around 10%. Optimize air flows, furnace
performance, and establish operating procedures to assure the
production of the desired end product.
3. Estimate and/or measure emissions from each source operation;
pellet lift and cooler emissions should be low and not require
control. If the hammermill collector discharge appears dirty,
consider controlling it with a bag house. Determine whether
visual opacity problems exist in the primary cyclone effluent.
It will probably require some form of source testing to determine
if the primary cyclone emissions exceed the standards.
A. Analyze potential production constraints; analyze the effects and
economic feasibility of reducing production during periods of low
hay quality. Operating the plant at a slightly reduced rate dur-
ing periods of low hay quality may be all that is necessary in
addition to step two to keep the plant operating in compliance.
Compare the resultant value of lost production with estimated
-------
cost of installing and operating add-on control devices.
5. Select suitable control systems; as required, select suitable
add-on control systems considering performance and cost trade
offs of alternative systems. Key points here are the
manufacturers' willingness to guarantee results in the operat-
ing constraints associated with each control system. One should
not design or buy control equipment without having a good measure-
ment of the gas flow it is supposed to handle.
The most cost-effective scheme for the control of particulate emissions
will most likely vary from plant to plant depending upon existing plant
equipment and available space.
-------
SECTION III
INTRODUCTION
Dehydrated alfalfa is the aerial portion of the alfalfa which has been
artificially dried under controlled conditions to insure maximum
integrity of nutrients. Chopped alfalfa is brought in from the field
and transferred to an automatic feeder that meters it directly into the
combustion gases in the drier. The drier may be either a triple-pass or
single-pass rotating drum in which lifting flights continuously raise
the alfalfa and drop it into the gas stream which moves it rapidly
through the drier in a concurrent manner. During dehydration, most of the
moisture of the alfalfa is removed in the constant rate drying phase
where moisture diffuses to the surface of the particles as fast as it is
evaporated from the surface. After remaining in the drum for about two
to ten minutes, the alfalfa (which is now quite dry), is separated from
the moisture laden gases by means of a cyclone or other separator. This
is called the primary cyclone or primary collector. The alfalfa (dry chop)
is then transferred either directly or through another cyclone for cooling
before going to a hammermill for grinding. The ground material is
usually pelleted, cooled and placed in Inert gas storage facilities.
Dryer generated emissions are discharged into the atmosphere from the
primary cyclone collectors which separate the dried alfalfa chops from
the dryer effluent. Hammermill and pellet cooler generated emissions
are discharged into the atmosphere from the hammermill and pellet cooler
collectors. The need to control dryer emissions has been established from
test results obtained from a 1971 American Dehydrators Association study(l)
which showed that emissions from the drying operation comprise more than
75% of total particulate emissions from a dehydrating plant. Whereas
the emissions from the hammermill and pellet cooler collectors are
generated by mechanical action and fall in the particle size range of
1 to 100 microns, the emissions from the primary collector are formed by
mechanical action and volatilization of organic matter from the alfalfa
on contact with the hot dryer gases.
The rapid absorption of heat by evaporation keeps the plant substance
cool enough (i.e. wet bulb temperature) to avoid burning as long as
there is moisture in the material. Were it possible to heat all of the
incoming wet material with perfect uniformity, the wet solids would never
reach a temperature exceeding the dryer outlet temperature. However, the incom-
ing stock is not uniform in cross sectional area or in moisture content.
For this reason, some of the incoming alfalfa (leaves or other small cross
sectional pieces) will dry completely while still in contact with gas from
the burner which has not been cooled by contact with a large amount of
water material. These parts will reach temperatures high enough to
evolve organic vapors or "smoke". Organic emissions (smoke) occur in all
types of processes where organic material is heated above 200° or 300° F.
This fume-forming process involves condensation of the vaporized organic
material as the temperature drops below the "dew point" temperature
for the material in question. The condensation produces very tiny particles
which may be as small as 0.01 microns in diameter at the point of formation.
These tiny fume particles grow very rapidly by electrostatic agglomeration.
-------
This mechanism becomes less effective as the particles become larger
and also as they become fewer in number. The result is that the particles
tend to stabilize rather quickly at sizes in the 0.2 to 0.5 micron range
depending upon initial concentration, residence time, and other factors.
In the summer of 1972, the American Dehydrators Association initiated a
program to assess the effectiveness of techniques for the control of
mechanically generated and fume formed particulate emissions from alfalfa
dryers. This involved field measurements of seven control devices or
systems - two pilot-scale and three full-scale wet collectors (scrubbers)
and two full-scale control systems which recycle effluent from the primary
cyclone. These devices/systems were suggested by the respective equipment
manufacturers to provide effective control
-------
Process conditions were monitored either by a process engineer from
MRI or by someone to whom that responsibility was delegated. Control
equipment operation conditions, i.e., gas-phase pressure drop and water
usage, were measured or estimated by representatives of the equipment
manufacturers.
Following the conclusion of the 1972 field testing, the equipment
manufacturers were asked to submit performance and cost data for a
"model" control problem. These data were analyzed to determine the
relative cost of purchasing, installing and operating each control
device system and the cost for water treatment, if applicable. Midwest
Research Institute personnel developed the model control problem and
prepared the cost analysis.
-------
TABLE 1
TESTING PROGRAM - EQUIPMENT AND PLANTS
Control
Manufacturer
Fisher-Klosterman (FK)
Applications
Corporation (APFCOR)
Koch Engineering Co.
Air Conditioning Corp.
(ACC)
Thompson Dehydrating Co.
Hell Co.
Equipment
Device/System
Wet Scrubber
Wet Scrubber
Wet Scrubber
Wet Scrubber
Wet Scrubber
Wet Scrubber
Wet Scrubber
Recycle System
Recycle System
Recycle System
Recycle System
Plant Modifications
M ii
ti it
Dehydrating Plant
Scale
Pilot
Pilot
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Full
Location
Neodesha, Kansas
Neodesha, Kansas
Oxford, Kansas
ii ii
Rozel, Kansas
Lawrence, Kansas
Lexington, Nebraska
Topeka, Kansas
St. Marys, Kansas
Grand Island, Nebr.
Dundee, Kansas
Abilene, Kansas
Neodesha, Kansas
Berthoud, Colorado
Source Operation**/
D+HM
D+HM
D
D
D
D+HM
D
D+HM+PM
D4-HM4-PM
D
D
D+HM+PM
DfHM
D+HM
Evap. Capacity
of Dryer (Ib/hr)
20,000
20,000
12,000 (1972)
19,000 (1973)
18,000
12,000
30,000
30,000
27,000
34,000
34,000
22,000
20,000
20,000
af Operations which contribute particulate matter to the primary cyclone effluent, i.e. drying (D), hanrnie milling (Hf
and pellet mill system (PM).
-------
SECTION IV
OBJECTIVES
The primary objective of this project was to determine the
feasibility of controlling the effluent from the primary
collectors of alfalfa dehydrating plants by recycling the
emissions or by using wet scrubbers. This includes several
related specific objectives. These were:
A. To evaluate effectiveness of recycling a portion
of the effluent gases back to the furnace inlet in:
1) reducing the amount of particulate emissions
to atmosphere; 2) increasing the thermal efficiency
of the overall system; 3) improving nutrient quality
of the dried product.
B. To determine the effectiveness of a system for
agglomerating, separating and recycling particulate
in the effluent gases back to the primary cyclone
and inlet to the furnace.
C. To measure the efficiency, horsepower requirements,
performance characteristics and associated problems
relative to the use of wet scrubbers for the
removal of particulate emissions from the effluent
gases of alfalfa dehydrating plants.
D. To test medium energy wet scrubbers with different
impingement mechanisms to obtain sufficient data
for evaluating the capability of wet scrubbers to
control emissions from alfalfa plants and to
determine if any particular type of wet scrubber
is more effective in the control of effluent from
alfalfa dehydrating plants.
E. To gather data relative to the effect of product
conditions (moisture and protein levels) and
process parameters (production rate, gas temperatures,
water spray in feeder) on particulate emissions from
alfalfa dehydrating plants.
F. To determine the degree of particulate emission
control which can be realized from plant modifica-
tions, air flow balance and operating procedures.
11
-------
SECTION V
DESCRIPTION OF CONTROL EQUIPMENT
This section presents a summary of the principles of operation
and a description of the plant interfacing for each of the ten
control systems/devices which were tested in 1972 and 1973. In
each case, a flow diagram is given which shows duct sizes, fan loca-
tions, approximate gas flow rates, and the locations of sampling
stations.
It should be noted that the control equipment operating principles
given below are based on wording, pictorial representations, and
diagrams provided by the respective equipment manufacturers; the
asserted particulate collection mechanisms are not necessarily
substantiated or refuted by the test data presented in Section VII
of this report.
Fisher-Klosterman Wet Collector (Pilot Scale)
Figure 1 presents a pictorial representation of the Fisher-Klosterman
wet collector, with annotations describing the principles of operation.
A pilot-scale F/K unit designed to handle about 700 acfm was installed
at Western Alfalfa Corporation's plant at Neodesha, Kansas. As shown
in Figure 2, a fraction of the skimmer recycle from the primary cyclone
collector to the furnace, was ducted to the F/K unit. Following the
initial testing, the position of the auxiliary fan was changed so that
the F/K unit operated under negative rather than positive pressure.
Also, a heater was added to raise the temperature of the scrubber
effluent gases above the dew point to facilitate testing at the outlet.
The F/K unit was supplied with tap water at about 4 gal./min. The
measured gas pressure drop was about 11 in. of water.
APPCOR Wet Collector (Pilot Scale)
Figure 3 presents a schematic diagram of the APPCOR wet collector.
The gas to be scrubbed enters from the top of the unit and flows
downward past a sawtooth overflow weir. The weir is used to intro-
duce make-up and/or recirculated water into the scrubber.
A scrubbing action is created by the gas shearing the water flowing
from the weir. The gas is then accelerated through an adjustable
venturi nozzle. By closing the nozzle, higher collection effic-
iency can be expected at the expense of an increased pressure
drop. At the nozzle tip, the dust-laden gas is impacted into the
sump, providing a second scrubbing action. The gas then takes a
135-degree turn, and at this point a large quantity of finely
divided water particles are sprayed up and throughout the interior
of the scrubber. The spray, which is caused by the accelerated
13
-------
Cleaned gas exhausts from
the outlet at low axial ve-
locity with little residual
kinetic energy.
The scrubbing liquid Is
Introduced at low pressure
Into the supply manifold
where it is distributed uni-
formly to the tangential
liquid supply tubes. The
design produces a contin-
wCii!- Him ol liquid on the
Interior surface of the col-
lector, which spirals down-
ward, welting the entire
collector wall, thus pre-
venting the formation of
wet/dry interfaces.
The upper section of the
collector It a highly effi-
cient centrifugal water
eliminator. As the cleaned
gas stream Is traveling
upward In a vortex, all
droplets are spun outward
to the liquid film on the wall,
where they are captured
and washed down to the
discharge.
The WL/WM Wet Dust Col-
lector Is a novel combination
of Venturi Contacting and
centrifugal water elimina-
tion. U. S. and Foreign
Patents have been applied
for.
The dust-laden gas enters the col-
lector Inlet.
The Venturi Contactor Is a precisely
shaped contraction In which the dust-
laden gas Is accelerated uniformly
to a high velocity. The throat of the
Venturi Intersects the body tan-
gentlally where the continuous water
Him Is shattered Into millions of tiny
droplets, each droplet capturing many
dust particles. These droplets reform
Into solid streams of dust-laden water
spiralling downward to the conical
sump.
The liquid, containing the collected
dust, flows Into the conical sump,
past the vortex breaker and out the
flanged discharge. From this point,
It may be piped to a settling pond or
clarifler tank for recirculatlon.
Figure I - Fisher-Klostennan Wet Collector (Courtesy of Fisher-Klostennan, Inc.)
14
-------
6" Dia.
10" Dia.
Skimmer Recycle
To Furnace
4,500acfm
Alfalfa Chop
Dryer Drum
Meal
Collector
Effluent
To Atmosphere
800 acfm
610 acfm
6)
FA
Unit
TESTS 1 AND 3
To Atmosphere
40,000acfm
a.
Primary
Collector
^
Dry Chop
F/K TEST 6
To Atmosphere
2,100 acfm
APPCORl
lUnit
To Atmosphere
2,100 acfm
S)
Figure 2 - Neodesha Plant Interfacing with Pilot-Scale Wet Scrubbers
15
-------
Clean Gas Out
Dust Laden Gas In
Optional
Outlet
Fresh
Water
Make-up
& Level
Control
Moisture
Eliminators
'VNfx/v?vrv:>l I i
--''''.' ''.''Y
Water Spray
Caused by Gas Impact
Sawtooth
Overflow
Weir
Recirculated
Scrubber
Water
Pump
Adjustable
Venturi
Nozzle
Figure 3 - APPCOR Wet Collector (Courtesy of Applications Corporation)
-------
gas shearing liquid from the sump as the gas changes direction,
provides the third scrubbing action. The gas is then passed
through a series of vertical chevron moisture droplet eliminators
and out the back or top of the scrubber.
A pilot-scale APPCOR unit was Installed at Western Alfalfa Corpora-
tion's plant in Neodesha, Kansas. As shown in Figure 2, a portion
of the skimmer recycle from the primary collector to the furnace, was
ducted to the scrubber unit.
The APFCOR unit was operated with recycled water which was supplied
from a water clarification system. The water recirculation rate
was estimated by the manufacturer to be 6 gal/min. The measured
gas pressure drop was about 7 in. of water.
Koch Wet Collectors
Figure 4 presents a pictorial representation of a Koch wet collector
containing three scrubber trays. Each tray contains numerous
venturi openings, and each opening is surmounted by a spider cage
holding a floating cap. In addition, each tray is equipped with
one or more downcomers and weir flow baffles that control the scrubbing
liquid as it flows across the tray and then to the tray below. The
particulate-laden gas enters the bottom inlet and flows upward through
the caps. The liquid flows across the deck and is kept in constant
froth by the gas which exits each cap. A head of frothy liquid is
maintained by the weir, providing intimate gas/liquid contact. The
tray divides the liquid flow into fine droplets that capture
particulates. By using multiple trays, division and redivision of
the scrubbing liquid occurs with high gas and liquid residence time
on each tray. Before the gas leaves the scrubbing chamber, it passes
through a mist eliminator.
A single Koch tray and a mist eliminator pad were installed in a
silo (10 ft diameter x 60 ft high) at Oxford Dehydrating Company's
plant in Oxford, Kansas. Effluent from the primary collector was
ducted to the base of the silo as shown in Figure 5. Recirculated
water was supplied to the tray from a settling basin at the bottom
of the silo. The manufacturer estimated the recirculation rate to
be 125 gal/min. and the fresh water make-up rate to be 2 gal/min.
The measured gas pressure drop was about 3 in. of water.
Two Koch trays and a mist eliminator were installed in a steel shell
at a plant in Rozel, Kansas, owned by Bert and Wetta Sales, Inc.
The connecting ductwork is shown in Figure 5. Recirculated water
was supplied to the trays from a settling basin at the bottom of the
collector. The manufacturer estimated the recirculation rate to be
175 gal/min. and the fresh water make-up rate to be 2.5 gal/min.
According to the manufacturer, the gas pressure drop was 6 in. of water.
The scrubber unit was removed from Rozel and installed at the Western
Alfalfa Corporation plant in Lawrence, Kansas for the 1973 tests.
One of the Koch trays was removed and the mesh type mist eliminator
was replaced with a chevron type. Pressure drop across the unit was
3-3% inches of water column.
17
-------
Figure 4 - Koch Wet Collector
(Courtesy of Koch Engineering
Company, Inc.)
-------
Dryer _
Effluent
1
Primory
Collector
To A
36" Di
©
(j) 22.500 ac
jpl.SOOacfm
1 36" Dia ._ Air Conditioning
L-T j*" * Corp. Unit
fnx
i
a.
^M
Fm
sphere
Dry Chop
Dryer
Effluent
LEXINGTON
38" Dia.
Dry Chop
OXFORD
Effluent
Dry Chop
ROZEL
To Atmosphere
Koch
troy in
silo
©
17,000acfm
-10' Dia. x
60' High
To Atmosphere
36" Dia.
1
Primary
Collector
33,200acfm
f 1 42" Dia.
L-Cs
(9)
29.400acfm
Koch_
trays in
steel"
shell
L 1
>9' Dia. x
18' High
*
\y
Figure 5 - Plant Interfacing for Full-Scale Wet Collectors
19
-------
Air Conditioning Corporation Wet Collector
The Air Conditioning Corporation wet scrubber consists of a
long rectangular duct through which gases flow horizontally. The
duct contains several banks of spray nozzles and the scrubbing
liquid is atomized by pumping the liquid under pressure through
the nozzles. At the downstream end of the scrubber is a mist
eliminator.
An ACC scrubber unit which was approximately 4 ft x 4 ft x 10 ft
was installed at a plant in Lexington, Nebraska, owned by Dawson
County Feed Products, Inc. The connecting ductwork is shown in
Figure 5 . Water was supplied from a fire hydrant to the unit
at a rate of approximately 405 gal/min based on the
manufacturer's nozzle performance calculations. According to
the manufacturer, the gas pressure drop was 3 in. of water.
Thompson Recycle System
Figure 6 presents a pictorial representation of the Thompson
recycle system. Effluent from the primary collector is passed
through two venturi sections separated by a static regain
chamber. The purpose of the venturi sections is to promote
agglomeration of fine particles. Water spray is provided in
the first venturi section to cool the gases and to increase
agglomeration. The gases in the static regain chamber are cooled
to saturation temperature by mixing gases with ambient air (the
damper on the air intake line was closed during testing; however,
leakage into the static regain chamber was observed). Gases
leaving the second venturi pass through the fan and into a
360-degree rectangular helicoid elbow. The largest particles
are skimmed from the effluent and returned to the primary collector;
Intermediate sized particles are returned to the front end of
the dryer drum (the return line to the drum is not shown in
Figure 6); and the smallest particles which are skimmed, are
returned to the dryer furnace.
In 1972, the recycle system was tested at the Thompson Dehydrating
Company's plant in Topeka, Kansas. At that plant, effluents from
the hammermill and pellet mill dust separators were returned to
the primary collector; thus, the plant had only one emission point.
A schematic diagram of the Topeka plant, showing the location of
sampling stations, is presented in Figure 7 .
The venturi spray used about 2 gal/min of water, according to
the manufacturer's estimate. The usage of water spray in the
feeder was measured at about 6 gal/min. According to the
manufacturer, the total gas pressure drop across the Thompson
recycle system was 9 in. of water.
The 1973 tests were conducted on the Thompson recycle system at
the Thompson St. Marys, Kansas plant. This system is essentially
the same as the one at Topeka. However, no water spray was used
in the Venturi or at the feeders during the St. Marys tests.
20
-------
HIGH ENERGY
SEPARATOR
Figure 6 - Thompson Recycle System (Courtesy of Thompson Dehydrating Company)
-------
To Atmosphere
34,000acfm
©
38" Dia.
Venturi-
Expansion-
Venturi
Chop
Meal
Figure 7 - Topeka Plant Interfacing With Thompson Recycle System
22
-------
Hell Recycle System
The Hell recycle system skims off a large portion (about 35%) of
the effluent from the primary collector and returns it to the
dryer furnace for incineration of the particulate matter. The
recycle duct is insulated to prevent heat losses and cooling of
the effluent below the dew point. No water is used in this system.
In 1972, the Heil recycle system was tested at a plant in Grand
Island, Nebraska, owned by Morrison and Quirk, Inc. A schematic
diagram of the Grand Island plant, showing the location of
sampling points, is presented in Figure 8. An elbow and horizontal
extension duct had been installed on the outlet of the primary
collector to provide for accurate measurement of the particulate
emissions. The extension duct was reduced in diameter, from 74 in.
to 60 in., to Increase the velocity within the duct so that accurate
flow measurements could be made.
The Heil recycle system at Dundee, Kansas which was tested in 1973
is basically the same as the system at Grand Island. The plant
flow diagram is shown in Figure 9.
To Atmosphere
Skimmer Recycle
34"Dia. To Furnace |21,000acfm
Alfalfa Chop
Dryer Drum
Primary
Collector
Dry Chop
Figure
8 - Grand Island Plant Interfacing With Heil Recycle System
23
-------
*FLOW RATES ARE PER TEST #140.
TO ATMOSPHERE
CO
JS
ESTIMATED 26,130 ACFM
ALFALFA CHOP
t>
HEILCO.
SD125-42 DRYER
TO ATMOSPHERE
26,770 ACFM
TO ATMOSPHERE
TO STORAGE
Figure 9. Dundee plant interfacing with Heil Recycle System.
-------
Abilene Plant Modifications
Several significant modifications were made on the alfalfa
dehydrating plant at Abilene, Kansas prior to the 1973 produc-
tion season. A new furnace and short flame burner (designed
to burn the gas in a three foot flame) were installed to
reduce the emissionswhich result from flame impingement on the
green chops. The primary cyclone system was converted from
positive to negative pressure air flow and was exhausted into
a large concrete (20' X 40') silo with sealed roof and 36"
diameter stack with necessary scaffolding and port holes for
source testing.
The hammermill and pellet mill systems were also made negative
and exhausted into the silo at the same point the primary cyclone
effluent entered the silo. A recycle line was installed from the
drum fan to the furnace with the capability of returning up to
30% of the gases back to the furnace. The Abilene plant inter-
facing is shown in Figure 10.
Lawrence Plant Modifications
This plant has two 8' X 24' - 3 pass alfalfa dehydrating drums.
Prior to modifications, the plant was of rather standard design
with positive primary collectors, secondary cooling collectors,
meal collectors, pellet lift collectors and pellet cooling collec-
tors.
The modifications consisted of a complete re-design and re-build-
ing of air flow systems from the drums and converting from
positive systems to one main negative system.
The two drums were ducted to a single negative primary cyclone.
The dehydrated chops drop from the cyclone into a pull-through
hammermill via a large rotary airlock. The hot gasses are drawn
from the top of the cyclone to the main mill fan which, in turn,
can exhaust directly to atmosphere or by means of a valve to the
Koch single tray scrubber.
The hammermill system employs a closed loop which returns all but a
small portion (20-30%) of the air back to the hammermill. A small
air balancing line, to prevent heat build up, is located between
the return loop to the hammermill and the suction side of the primary
mill fan. Figure 11 presents the plant flow diagram.
25
-------
fo
MAX. 30% AIR
TO FURNACE
Figure 10 Abilene plant interfacing.
-------
MEAL MEAL
COLLECTOR COLLECTOR
Figure 11. Lawrence plant interfacing.
-------
Neodesha and Berthoud Plant Modifications
The Neodesha, Kansas and Berthoud, Colorado plants have similar
flow diagrams shown in Figure 12. Both have positive pressure
primary cyclone collectors and modified hammer-mill meal handling
systems. Careful attention was given to balancing air flows and
sizing of collectors.
A positive displacement skimmer was installed at both plants
between the top of the primary collector and the intake to the
drum furnace handling about 6000 CFM on a continuous basis. At
Berthoud, the skimmer entry device to the furnace was modified to
give better distribution of the recycled gases into the furnace.
The furnace was also lengthened an additional three feet in an
effort to reduce flame impingement on the green chops entering
the drum.
28
-------
RECYCLE LINE
TO FURNACE
N>
vO
TO ATMOSPHERE
MEAL MEAL
COLLECTOR COLLECTOR
H
COOLER PELLET ASPIRATOR
COLLECTOR COLLECTOR
10 x 40 THREE PASS
DRYER DRUM
ALFALFA CHOP
Figure 12. Neodesha and Berthoud plants interfacing.
-------
SECTION VI
MEASUREMENT TECHNIQUES, EQUIPMENT AND PROCEDURES
Two types of source tests were performed to determine, respective-
ly; the mass flow rate of particulates, and the particle size
distribution. The mass rate test consisted of the measurement of
dust loading and the carrier gas flow rate, temperature and
composition (by moisture and Orsat analysis)
Integrated particulate samples, representative of the entire duct cross
section, were collected by sampling for equal intervals of time over a
network of properly distributed points. For each test, the duration
of sampling was sufficient so that short term fluctuations in
particulate flow were averaged.
For the particulate sizing tests, an Andersen in-stack impactor was
mounted on the end of the sampling probe in place of the normal tip.
The Andersen impactor measures size distribution in situ, thereby
eliminating particle agglomeration problems encountered when
particulate samples must be transferred before sizing analysis.
Table 2 indicates the process parameters that were measured
during testing and the frequencies and methods of measurement.
These parameters have been classified into three groups: raw
materials, product (pellets), and process operating conditions.
In most cases, samples of alfalfa, i.e., green chops, dry chops,
meal and pellets, were sent to testing laboratories for moisture
analysis. The laboratory results were considered more accurate
than results of analysis in the field with a portable moisture
balance. Therefore, field results are given in this report only
for the cases for which laboratory results were not obtained.
Isokinetic sampling was used to collect particulate samples which
validly represented average conditions (particulate concentration
and particle size distribution) over the entire cross section of
the flow stream. The sampling train is described in the Federal
Register(2) and in a recent paper by Cowherd and Vandegrift(3).
An s-shaped pitot tube is attached to the probe so that velocity of
the flow stream near the probe tip may be monitored while the sample is
being withdrawn. In this way, rapid adjustment of the sampling rate
can be made to attain the isokinetic condition.
31
-------
Parameter
TABLE 2
PROCESS PARAMETERS MONITORED DURING TESTING
Units Measurement Frequency
Measurement Method
M
I. Raw Materials
A. Hay (green chops)
1. Moisture Content
2. Feed Rate
B. Fuel Consumption Rate
II. Product (pellets)
A. Protein Content
B. Moisture Content
C. Production Rate
III. Operating Conditions - Internal
A. Dryer Conditions
1. Outlet Temperature
2. Excess Air
B. Moisture in Hanmermill Feed
(dry chops)
C. Recycle Flow Rates
percent by weight
ton Air
scfm
percent by weight
percent by weight
ton/hr
"F
percent
percent by weight
acfm
composite of 3 samples/test
beginning and end of test
composite of 2 samples/test
composite of 2 samples/test
3/hr
3/hr
continuously integrated sample
composite of 3 samples/test
Cenco balance/Lab determination
Truck weights
Meter reading
Lab determination
Lab determination
Scale dumps
Mercurial thermometer
Orsat analysis
Cenco balance/Lab determination
Duct velocity profile
-------
The average velocity was determined by measuring the flow over the
proper network of sampling points within the cross section of
the duct. The average temperature and moisture content of the
carrier gases were also measured with the standard particulate
sampling train. The dry-gas composition of the carrier gases was
measured by collecting an integrated gas sample in a tedlar bag
and by analyzing it with an Orsat apparatus.
Several difficulties inherent in testing the kind of effluent
produced by alfalfa dryers should be pointed out. Foremost
among these is the problem of measuring gas flows and grain
loadings for very high moisture content streams. The measure-
ment of velocities using the pitot-tube is not particularly
difficult, but the measurement of moisture content by condensa-
tion and collection of the liquid water seems to cause a fair
amount of difficulty. In several of the tests, the moisture
content is reported as considerably above the saturation level
for the sample temperature. This suggests that some of the
tests may have been run on a stream that was cooled below
the dew point and had condensation taking place in the
ductwork. Also, there are some difficulties associated with
measuring particle size when the gas stream is extremely wet,
and there is a possibility of condensation occurring in the
sampler. The problems are associated again with potential
volume change due to water condensation, and with the possibil-
ity of collected particles which have been wetted by water
droplets or agglomerated by water condensation.
The inherent inaccuracies of the methods used in this testing
program to measure gas flow rate, particulate concentration,
and particulate mass flow rate (the product of the two) have
been thoroughly analyzed in a recent paper by Shigehara et al.(4
The cumulative error in the measurement of flow rate is +5% or
less; in the measurement of emission rate, the cumulative error
is ±10% or less. However, there is the possibility of a 20%
error in the measurement of particulate rate in the presence
of flow swirl; this is because an S-shaped pitot tube may
read as much as 15% higher if sufficiently inclined relative
to the flow direction(5 ). Errors in the measurements
of process parameters are estimated to be +10% or less.
Procedures which were followed in the processing and analysis
of the collected particulate matter are essentially the same
as those outlined in the Federal Register(2 ). At the end
of each source test, the particulate sample was transferred
from the sampling train to appropriate containers and returned
to the Institute for precision laboratory analysis.
For the particulate sizing tests, an eight-stage Andersen
impactor was used in place of the normal probe tip on the
isokinetic sampling train. The impactor separated the
particulates into eight size classes with the greatest
resolution in the 0.5 to 15 u size range. The sampling time
for a sizing test was determined by the dust loading in the
effluent gases.
33
-------
Calculation of the results of the particulate emission tests was
done by electronic computer. Corrections were made for deviations
from isokinetic sampling; such deviations were generally less
than 10% and corresponding corrections less than 5%.
The equations for process weight rate (FWR) and evaporation rate
(ER) are as follows:
(100-MJ
PWR(lb/hr) = PR x 2000 ( _ J
(100-Mg)
(Mg -
ER(lb/hr) - PWR (___
(100-Md )
< 100-MJ (Mg -Md)
+PR x 2000 ( _ _) ( _ ) ;
(100-Mg) (100-Md)
Where PR = pellet production rate (ton/hr) ,
Mg = percent moisture in green chops (dryer input) ,
Md = percent moisture in dry chops (dryer output) , and
Mp = percent moisture in pellets.
For the 1971 and 1972 tests on the positive pressure cyclone
collectors, a 90° elbow and long horizontal extension duct were
installed on the cyclone outlet for the purpose of eliminating
the normal swirl of gases being discharged from the collectors.
Testing was conducted eight duct diameters downstream from the
elbow, but there was still occasional swirl and the long exten-
sion duct created some slight problems of back pressure and
settling of particulate natter. The cost of fabricating and
installing the long extension ducts was also very expensive,
running up to $2,000.
To overcome the difficulties associated with the extension duct,
two types of straightening vane arrangements were tried in 1973.
Both proved satisfactory, but the one shown in Figure 13 was
slightly preferable. Vertical extension ducts were used but,
with the swirl removed by the straightening vanes, it was possible
to use much shorter ducts and still be consistent with the Federal
Register (2) by sampling at the prescribed number of traverse points.
The sampling platforms and ladders were arranged basically as
shown in Figures 14 and 15. In designing straightening vanes,
each cubicle should be 5-15% of the stack area. Smaller cubicles
have less swirl, but more back pressure and vice versa. The
cubicle length should be three times the width of the cubicle.
Thus, smaller cubicles will lower the height of the straightening
vane section.
34
-------
TOP VIEW HONEYCOMB
, SIDE VIEW
i ONE CUBE
L=3W
EACH CUBICLE APPROXIMATELY
5-15% OF STACK AREA
Figure 13 Straightening vanes
STACK
*PORTS - 4 in. NIPPLE (4 in. TO 6 in. LONG) WITH CAP.
Figure 14. Testing platform.
35
-------
STACK
iJ
2-8 STACK
DIAMETERS
jf 'K!i'-r-
10 in. mm. J
1 /
TO ACCOMMODATE
STACK SAMPLING
a
I
31
1
42 in.
U MINIMUM OF 2 STACK DIAMETERS ABOVE STRAIGHTENING VANES. EIGHT STACK
STACK DIAMETERS ABOVE LAST DISTURBANCE IF STRAIGHTENING VANES
NOT USED.
2) VERTICAL PLATFORM HAND RAIL BRACES NOT OVER 2-5 in. FROM EACH
END OF THE 13 in. GAP IN THE HAND RAIL TO HELP SUPPORT TEST
EQUIPMENT.
Jj MIDDLE RAIL REMOVED AT WHICHEVER END OF THE PLATFORM
THE LADDER IS INSTALLED.
Figure 15. Testing platform sideview.
36
-------
SECTION VII
MEASURED PERFORMANCE DATA AND CONTROL EFFECTIVENESS
This section presents a summary of performance data and particle size
measurements and discusses the effectiveness of the emission control
devices/systems that were tested. The measures used to assess the
comparative effectiveness of the equipment are: the Bay Area process-
weight-rate emission standard; collection efficiency versus particle
size for the wet collectors; and overall collection efficiency for the
recycle systems.
Fisher-Klosterman Wet Collector
Table 3 presents a summary of measured performance data for the Fisher-
Klosterman pilot-scale wet collector at the Neodesha plant. The
particulate samples obtained during Tests 1 and 3 indicated the
possibility of contamination of the outlet stream with particulate mat-
ter which had condensed In the ducting near the fan during periods when
the F/K unit was not operating. This would have the effect of decreas-
ing the apparent collection efficiency. The ducting was redesigned
and the air and water flow rates were adjusted for Test 6 with the
resulting improvement in the collection efficiency.
It should be noted that the average particle size at the inlet to the
F/K unit was much smaller than the size measured in the cyclone
effluent at the Neodesha plant during the 1971 growing season. The
particulate loading was also considerably reduced from that which was
measured in 1971. This may be due to the long length of small diameter
duct which transported the gases to the F/K unit. It appears likely
that some of the large particles were deposited on the Inside surfaces
of this connecting ductwork.
Figure £ presents the particle size distribution obtained at the
inlet to the F/K collector. The fine particle size and the low grain
loading at the collector outlet made it impossible to obtain an
acceptable size distribution measurement.
Based on the average of three tests, the F/K collector reduced particu-
late emissions by 46%. However, the reduction in emissions on Test 6,
after the system was modified to eliminate condensation, was 58.6%
as compared to an average of 40% on Tests 1 and 3.
APFCOR Wet Collector
Table 4 presents a summary of measured performance data for the APPCOR
pilot-scale wet collector. During testing of the APPCOR unit at the
Neodesha plant, an attempt was made to increase emissions. This
apparently resulted in considerably larger particle size at the inlet
to the APPCOR unit than had been measured in the 1971 tests of the
effluent from the primary collector at the Neodesha plant.
Figure 17 presents the particle size distributions that were obtained
at the scrubber inlet and outlet.
Based on the average of three tests, the APPCOR unit reduced particulate
emissions by 64%.
37
-------
LO
oa
Fisher-Klosterman Wet Collector
Scale Pilot
Approx. Water Usage = 4 gpm
Approx. Pressure Drop 11 in. H
Pellets
Test
1
3
6
Prod. Rate
(tons/hr)
2.85
2.62
3.00
APPCOR Wet Collector
Scale: Pilot
Approx. Water Usage 6 gpm§'
Approx. Pressure Drop = 7 in. H20
Pellets
Test
2
3
4
Prod. Rate
(tons/hr)
2.75
1.95
2.15
TABLE 3
FISHER-KLOSTERMAN PERFORMANCE DATA
Green Chops
Moisture
74.3
77.9
71.2
Green Chops
Moisture
(7.)
76.8
77.3
73.8
Collector Inlet Conditions
Flow Rate Partic. Loading Avg. Particle
(acfm) (Rr/acf) Size (u)
859 0.0617 < 1
751 0.0818 < 1
672 0.0497 < 1
TABLE 4
APPCOR PERFORMANCE DATA
Collector Inlet Conditions
Flow Rate Partic. Loading Avg. Particle
(acfm) (Rr/acf) Size (n)
2,230 0.0879 11.0
2,010 0 1160
2.100 0.1670
Plant Neodesha
Source Operation. D+HM
Dryer Capacity = 20,000 Ib/hr
Collection
Partic. Rate Efficiency
(Ib/hr) (%)
0.45 37.8
0.53 41.5
0.29 58.6
Plant Neodesha
Source Operation tH-HM
Dryer Capacity = 20,000 Ib/hr
Collection
Partic. Rate Efficiency
(Ib/hr) (%)
1.67 54 5
2.01 67 2
3.02 65.2
a/ Estimated recirculation rate
-------
100.0
MM Ml Ml
10.0
2
U
1.0
y
I
o.io
o.oi
4:
I
fcu
WEIGHT % GREATER THAN STATED SIZE
H W M » J 1 05 OT 0.1 ».0» Ml.
} Fisher- Klosterman Wet Collecto
Scale: Pilot
Plant: Neodesha
Source Operation: Dryer + HM
-
-r
-
H
111 0.0f 0.1 07 01 1 I t M
-i--f
r-r
I
f
r-
il
t
I
WEIGHT % LESS THAN STATED SIZE
Figure 16- Particle Size Distributions - F/K Wet Collector
39
-------
»tM Ml Ml
WEIGHT % GREATER THAN STATED SIZE
M 40 M 10 Id I 3 I at t.l 9.1 0.0>
t/i
I
Q
U
Test 2 (Outlet)
APPCOR Wet Collector
Scale: Pilot
f Plant: Neodesha
Source Operation: Dryer + HM
0.01
Ml M.f M VI
WEIGHT % LESS THAN STATED SIZE
Figure 17 _ Particle Size Distributions - APPCOR Wet Collector
40
-------
Koch Wet Collectors
Table 5 presents a summary of 1972 performance data for the single
Koch tray that was installed in the silo at the Oxford plant. The
measured partlculate loading and average particle size at the
inlet to the silo was fairly typical of values for the primary
cyclone effluents tested in 1971. Figure 18 gives the measured
particle size distributions.
At the Oxford plant, the single tray Koch scrubber, based on
the average of two tests, reduced particulate emissions by 57%.
The Oxford plant was operated between 67% and 94% of capacity
and had uncontrolled emissions of about 70% of the allowable
limit of the Bay Area process-weight-rate standard.
Table 6 presents a summaiy of performance data for the two-tray
Koch unit at the Rozel plant. When the Koch unit was connected,
a deficiency in fan capacity and resultant decreased air flow
caused the furnace to cycle from idle to maximum output. This
apparently resulted in the generation of additional fine
particulate matter from the burning of the alfalfa and lowered
the average particle size at the collector inlet. (For Test 6
the scrubber unit was disconnected, and the emission rate was
substantially reduced.) This operating problem introduces
uncertainty into the validity of the results obtained at this
plant and may have accounted for a measured efficiency below that
of the single tray Koch scrubber tested at Oxford.
Figure 19 gives the particle size distributions that were obtained
at the Rozel plant.
The two tray unit at Rozel, based on the average of three tests,
reduced particulate emissions by 26.5%. This brought the plant
into compliance with the Bay Area standard because the uncontrolled
emissions were fairly close to compliance.
This dryer was operated above its rated capacity based on 75%
moisture in the green chop and operated within compliance of the
Bay Area standard on the one test made with the Koch unit
disconnected.
Table 7 presents a summary of the 1973 data obtained on the Koch
units at Oxford and Lawrence, Kansas. Efficiencies are not shown
for these tests because simultaneous inlet and outlet measurements
were not taken. The purpose of these tests was to measure only
the degree of control which could be obtained with the wet scrubber;
not the scrubber's efficiency.
41
-------
TABLE 5
KOCH (OXFORD) PERFORMANCE DATA
Koch Wet Collector
Scale: Full
Approx. Water Usage - 125 gpm4
Approx. Pressure Drop 2.8 In. H.O
Pellets Green Chops
Prod. Rate Protein Moisture
Test (tons/hr) «) rt)
1 2.07 20.9 80.3
3 1.49 18.3 81.0
£/ Estimated reclrculatlon rate; estimated make-up rate
Koch Wet Collector
Scale: Full
Approx. Water Usage 175 gpn£'
Approx. Pressure Drop 6 in. H20
Pellets Green Chops
Prod. Rate Protein Moisture
Test (tons/hr) (1) (I)
1 3.94 18.3 74.6
2 4.31 - 71.9
4 3.46 20.1 76.4
6 4.22 18.6 59.3
Collector Inlet Conditions
Flow Rate Partic. Loading Avg. Particle
(acfm) (xr/acf) Sice GO
19,300 0.0780 6.0
18,600 0.0673
2 gpm.
TABLE 6
KOCH (ROZEL) PERFORMANCE DATA
Collector Inlet Conditions
Flow Rate Partic. Loading Avg. Particle
(acfm) (Rr/acf) Sice GO
32,100 0.0986
33,100 0.1075
34,300 0.0674 < 1
42,600 0.0370 2.8
Plant : Oxford
Source Operation
Dryer Capacity
Partic. Rate
(Ib/hr)
12.9
10.9
Plant: Rocel
Source Operation
Dryer Capacity
Partic. Rate
(Ib/hr)
27.1
30.5
19.8
13.2
: Dryer
12,000 Ib/hr
Collection
Efficiency
tt)
62.0
52.6
: Dryer
18.000 Ib/hr
Collection
Efficiency
19.2
23.3
36.9
a/ Estimated reclrculatlon rate; estimated make-up rate 2.5 gpm.
-------
100.0
WEIGHT % GREATER THAN STATED SIZE
I
u
I
2
Koch Wet Collector
Scales Full
Plant i Oxford
Source Operation* Dryer |
a£:J.-:i» ::ii:S = = s:|.!ii t =': f ftti Uif t It Htt tffi: ffi ;!i;Sp=
0.01
Q 1 01 04 I 7 t It M»«DbOU70W
WEIGHT % LESS THAN STATED SIZE
11 *! M.W
Figure 18 - Particle Siz«f Distributions - Koch Wet Collector (Oxford)
43
-------
WEIGHT % GREATER THAN STATED SIZE
100.0
in M.I MJ
M .! Ml !
10.0
z
o
Qi
U
u
fe
Scale: Full
Plant: Rozel
Source Operation: Dryer
WEIGHT % LESS THAN STATED SIZE
Figure 19 - Particle Size Distributions - Koch Wet Collector (Rozel)
L.L.
-------
U1
TABLE 7
KOCH (OXFORD) 1973 PERFORMANCE DATA
Pellets
Test
104
105
109
110
111
112
113
130
131
132
146
147
148
Prod. Rate
(tons/hr)
2.83
3.17
2.75
2.50
2.55
2.50
2.80
2.45
2.45
2.60
2.65
2.42
2.53
Protein
(%)
18.4
18.5
17.0
17.0
17.0
18.6
18.6
16.3
16.3
17.6
17.8
18.8
19.4
Chops Moisture %
Green Dry
75.0 9.85
75.0
KOCH
78.6
80.1
77.5
81.5
70.4
76.6
77.5
76.8
72.0
76.8
73.4
9.2
(LAWRENCE)
14.6
17.4
16.7
14.9
12.6
16.9
14.0
15.0
10.1
9.7
11.9
Source
Operation
D
D
PERFORMANCE
D+HM
D+HM
D+HM
D+HMi./
D+HMi/
D+HM
D+HMl/
D+HMl/
D+HML/
D+HMi/
D+HM
PWR
Ibs/hr
20,800
23,300
DATA
23,400
23,400
21,100
24,800
17,800
19,500
20,500
20,800
18,100
20,000
18,200
Dryer Capacity =
Evap.
Rate Far tic. Rate
Ibs/hr (Ib/hr)
15,100
16,900
Dryer
17,500
17,800
15,400
19,400
11,800
14,000
15,100
15,200
12,500
14,800
12,700
9.2
11.4
Capacity =
22
21.7
18.8
8.30
6.59
28.5
13.3
12.2
13.14
13.60
41.21
19,000 Ibs/hr
Visual
Opacity
12,000 Ibs/hr
< 10
< 10
JL/ With the wet scrubber in operation. On the other tests, the wet scrubber was by-passed.
-------
During the 1973 tests at Oxford, the effluent from the primary cyclones
for both drums was ducted to the silo containing the Koch Tray. The
plant was well in compliance with the Bay Area Standard when operated
up to 90% of evaporative capacity on slightly above average quality
material (18.5% protein hay). The hammermill system was controlled with
a bag filter which allowed only 0.01-0.03 Ibs. of particulate emissions
to escape per hour.
The plant at Lawrence, Kansas was tested both with and without the 2-tray
Koch scrubber. These tests were not made simultaneously, but Indicate a
collection efficiency of 57% for the wet scrubber since the average
uncontrolled emission rate was 26.44 Ibs/hr with an average controlled
emission rate of 11.28 Ibs/hr. The plant was either in compliance or
very close to compliance on all but one of the tests without the wet
scrubber. Tests 146-148 were run at higher back-end temperatures and
lower dry chop moistures than the previous tests In order to measure
the effectiveness of the wet scrubber and the level of uncontrolled
emissions under these conditions. The uncontrolled emissions increased
substantially, but the controlled emissions remained virtually unchanged.
The mesh type mist eliminator used at Oxford and Rozel was susceptible to clog-
ging. This problem apparently did not exist with the chevron type mist
eliminator used in the scrubber at Lawrence, Kansas.
Air Conditioning Corporation Wet Collector
Table 8 presents a summary of performance data for the Air
Conditioning Corporation wet collector at the Lexington plant.
The measured particle size distribution and particulate loading
at the collector inlet were typical of values for the primary
cyclone effluents tested in 1971.
The Air Conditioning Corporation system reduced the primary
cyclone emissions at the Lexington plant to a level well below
the Bay Area standard. Based on the average of three tests,
this unit reduced the particulate emissions by 79%.
It is likely that the measured collection efficiency of the Air
Conditioning scrubber was substantially increased because of the
large percentage (59-71%) of water vapor which condensed in the
scrubber. The condensation of water vapor was greatly enhanced
by the temperature of the water which was drawn from a nearby
fire hydrant. Thus, these results do not necessarily reflect
the efficiency of this unit under normal installation conditions
where the water would be recycled. Recycled water becomes heated and
effects very little condensation.
The Lexington plant had rather high uncontrolled particulate
emissions for a plant operating at only about 40% of capacity.
Figure 20 Presents the particle size distributions that were
obtained at the ACC scrubber inlet and outlet.
46
-------
TABLE 8
AIR CONDITIONING CORPORATION PERFORMANCE DATA
Air Condi t Jon ing Corp.
Scale:
Approx.
Approx.
Full
Water Usage -
Pressure Drop
, Wet Collector
405 gptn
3 In. H20
Pellets
Test
2
3
4
Prod. Rate
(tons/hr)
2.35
1.95
2.15
Protein
ffl
24.3
24.5
24.5
Green Chops
Moisture
(I)
76.3
82.3
82.3
Collector Inlet Conditions
Flow Rate Partic. Loading Avg. Particle
(acfm) (Kr/acf) Sice (p)
29.300 0.1434 4.8
32,200 0.0926
32,900 0.1331
Plant: It
Neodesha:
Dryer Cap;
Partic. Rate
(Ib/hr)
36.0
25.6
37.5
30,000 Ib/hr
Collection
Efficiency
72.8
73.8
91.5
-------
100.0
WEIGHT % GREATER THAN STATED SIZE
Ml I 1 i.l U II «
z
o
of
U
10.0
|j| Test 2 (Outlet)
Hit
Test 2 (Inlet
1 U
; Air Conditioning Corp. Wet Collector M
0.10
0.01
M M.I H.t
O.I O.I 0.1 1 t » » JO M 40 M
WEIGHT % LESS THAN STATED SIZE
Figure 20 - Particle Size Distributions - ACC Wet Collector
48
-------
Particle Size Distributions
The test results show particle size distributions from the
one micron and larger size range, which includes the
mechanically generated dusts, down into the sub-micron range,
which almost certainly contains fumes formed by volatilization
of part of the charge material. Another clear-cut indicator of
the combination of a dust and fume in a gas stream is a clearly
non-linear plot of cumulative frequency versus particle size
on log-normal probability paper. The plotted curves most often
show this statistically for the dryer effluent, although some of
the other gas streams show relatively linear distributions.
Theoretically, any mechanical grinding or attrition process should
produce dust which distributes in a log-normal manner. This is,
a plot of the logarithm of the particle diameter versus the
fraction in each size range should give an approximately normal
distribution curve.
Whenever particles generated by two different mechanisms which
have widely different average particle sizes are mixed together,
they form a bi-modal particle size distribution when plotted on
ordinary coordinates and when plotted on log-normal cumulative
probability paper they make a curve very much like the ones
shown in several of the plots.
Fractional Collection Efficiency
In assessing the performance of each wet collector as compared to
the others, the fractional collection efficiency is a more meaning-
ful measure than the overall collection efficiency, because the
particle size distributions at the scrubber inlet vary from plant
to plant. Figure 21 presents the fractional efficiency curves for the
wet collectors, which were calculated from the inlet and outlet
particle size distributions. Although it was not possible to plot
a curve for the Fisher-Klosterman unit due to the lack of an
acceptable size distribution measurement at the scrubber outlet,
its overall performance in collecting fine particles was favorable.
Figure 21 shows that the APPCOR scrubber unit was most efficient
in collecting the larger particles (approximately 20-microns)
although the collection efficiency of this unit decreased rapidly
as the particle size decreased. For the smaller particles (approxi-
mately 2-microns) the Koch unit at Oxford was the most efficient
followed by the Koch unit at Rozel and the Air Conditioning unit.
These collection efficiencies are within the accepted range for
medium efficiency wet scrubbers.(6)
However, the efficiencies obtained did not correlate well with the
scrubber pressure drop. The fundamental problem in relating the
scrubber efficiencies to the test conditions under which they
were obtained probably lies in the fact that the scrubbers are all
basically well-equipped to collect the large particles, but not
capable of doing very much of a job on the submicron fraction.
49
-------
99.90 -
99.99
1.5
A KOCH ( ROZEL )
V AIR COND.
O KOCH ( OXFORD )
D APPCOR
4 5 7 10
Particle Size ( Microns )
Figure 21 * Fractional Collection Efficiency Curves
50
-------
Summary of Wet Scrubbers
Wet scrubbers are generally grouped in three more or less arbitrary
ways. Low energy scrubbers or air washers operate at pressure drops
on the order of 1 or 2" water column and are effective in collecting
large particles (over two or three microns) at high efficiency.
They are not useful for collecting fumes or fine dusts, and are
not likely to be satisfactory for the alfalfa dryer application.
Medium energy scrubbers usually operate from 2 or 3" water column
to 10" or so, and are quite effective for all particulate matter
except submicron particles.
There is every indication from the test results, that a scrubber
with several inches of water pressure drop will bring the
particulate emissions down by at least 50% and will probably bring
a plant into compliance if the problem consists principally of
dust rather than smoke emissions. Several scrubber types ought to
be suitable. These include the multi-Venturi type, conventional low
pressure drop Venturi scrubbers, wetted high velocity cyclones, etc.
In addition, there is some indication that a wet scrubber is helpful
even when smoking is a significant problem. Although the scrubber
is not likely to be highly efficient on the sub-micron smoke particles,
it does do a good job of cutting the accompanying dust emission down,
and may improve the appearance of a smoky plume considerably.
High energy scrubbers with pressure drops ranging from 10" to as
high as 120" water column are frequently used to collect fume-like
materials from such sources as BOF steel making furnaces, glass
melting furnaces and coal dryers. While Venturi scrubbers at
pressure drops on the order of 60" water column are quite effective
in capturing submicron fume particles from smelting applications,
etc., they are often inadequate for the collection of organic fumes.
This is probably due more to the low specific gravity of the fume
material than to a large difference in particle size. The former
are usually some 2.5 times as dense as water, while the organic
fumes tend to be about the same density as water, or a little less.
The probability is great that a high energy Venturi scrubber would
have to operate at pressure drops between 60 and 100" water column
in order to reach collection efficiencies on the order of 90%
or so of the fume materials in the alfalfa dryer discharge.
51
-------
Water Clarification
The problem of water clarification and sludge disposal is inherent
to the use of wet scrubbers. At the Oxford Dehydrating plant, Oxford,
Kansas, this problem was solved with the following arrangement. The
bottom fourteen feet of the 10* X 60* concrete silo was used as a
storage tank for the water which were circulated over the Koch tray
in the silo. The bottom of the storage tank has a cone shaped hopper
with a drain at the lowest point.
When the scrubber is in operation, ten gallons of water sludge is
drained off per minute and pumped to a settling and evaporation
pond. This pond is waterproof and after the sludge settles to the
bottom, excess water is pumped to four rain bird spray nozzles,
irrigating a grass sodded area.
There is no noticeable odor throughout this system when only the
effluent from the primary collector is handled by the wet scrubber.
Approximately once a year the collection pond is drained, the
sludge is removed and spread on an alfalfa field for fertilizer.
A pilot commercial water clarifying unit was used for a few days
in 1973 at the Oxford Dehydrating Co. plant. Feed rate to the
unit was measured at slightly over 10 gallons per minute.
Chemical treatment consisted of the addition of lime and a synthetic
polymer, Drewfloc 260. The lime was added at a rate of approximately
3 Ibs per hour to produce a pH of 7. The polymer was added at the
rate of 1 lb/10 hrs. The cost of this chemical treatment was
roughly 25C per hour.
A clear effluent was produced almost immediately after chemical
treatment was initiated. The amount of solids removed indicates
the necessity for desludging the unit every eight hours. The
volume of concentrated sludge, while subject to variation from
operation to operation, would probably be about l/20th of the feed
volume, or 240 gallons every eight hours.
As an alternative to using lime as the alkali source, it may be
beneficial to use something else which would produce a more
soluble salt, such as sodium hydroxide or soda ash. Spot
checks run on beakers full of the clarifier feed indicate that the
primary function of lime is to adjust pH to a point where the floc-
culent or polymer would be effective. If a sodium based alkali is
used rather than lime, there would be less chance of developing a
saturated solution with respect to calcium salts in the scrubber
itself. The need to use a sodium based alkali would have to be
evaluated on an individual basis dependent upon the calcium hardness
of the make up water to the scrubber.
A flotation type water clarifying unit was designed, constructed
and installed by Western Alfdfa Corp. personnel at the Lawrence
plant. This unit worked quite satisfactorily, during the wet
scrubber tests, and did not require the use of chemicals to
coagulate the solids.
52
-------
Several corrosion problems have been encountered and should be
taken into consideration whenever anyone uses wet scrubbing
equipment for the collection of organic dusts. The water becomes
acidic and will attack carbon steel and possibly concrete over a
long period of time. The alternative solutions to this problem
involve the use of corrosion resistant materials such as rubber,
plastics, and stainless steel, or the controlled addition of lime
or caustic to keep the water in the scrubbing system alkaline.
Chemical control is probably the cheapest, but it requires very
good operating attention. Even a few days of operation with
acidic water is likely to damage carbon steel tanks, piping and
pumps.
53
-------
Thompson Recycle System
Table 9 presents a summary of 1972 performance data obtained for the
Thompson recycle system at the Topeka plant. During the first week
of testing (Tests 1-8), particulate loadings in the primary cyclone
effluent and particulate emission rates were high, compared to aver-
age values measured (at other plants) in the 1971 study. There was
also considerable evidence of the depositing and sloughing of particu-
late material from the inside of the ductwork. In order to reduce this
problem, the manufacturer decided to insulate the fan, helical separator,
and the outlet stack.
During the second week of testing (tests 9-12) particulate loadings
and emission rates were considerably lower than during the first week,
and the percent of particulate recycled was higher. Also, hay quality
was higher. Test 11 was run with the recycle lines shut off and
indicated little, if any, particulate buildup in the ductwork. However,
flow plugging problems occurred frequently, mostly at the hammermill.
After the testing had been completed, the manufacturers reported find-
ing equipment malfunctions which they feel substantially impaired plant
operations and particulate emission control.
Figure 22 presents the particle size distributions obtained at the
outlet of the primary collector and at the inlet to the fan (locations
one and five in Figure 7). These distributions show slight evidence
of particulate agglomeration in the venturi sections and the static
regain chamber. However, it should be noted that the outlet and inlet
samples were taken on different days with different hay quality and
production conditions.
The Thompson plant at Topeka was in compliance with the Bay Area
standard when operated at less than 75% capacity. At the higher
production rates (75% to 85% of capacity), the effectiveness of the
recycle system was reduced. The use of water sprays in the feeder
(tests 1, 8, 9, 10 and 12) may have reduced particulate emissions,
however, this effect cannot be isolated from the effect of reduced
process-weight-rate. During Test 11, the plant was nearly in compli-
ance without recycle and without water sprays. During Test 8, a higher
than normal water usage rate at the feeder was used to simulate a "wet"
hay condition which occurs as a result of heavy dew or rain.
The 1973 test data, Table 10, obtained on the Thompson recycle system
at St. Marys, Kansas show the plant to be well within compliance with
the Bay Area standard even when operated above 100% of rated evaporative
capacity. This is a definite improvement over the results obtained
under the conditions of the 1972 Topeka tests. However, it is
difficult to make a direct comparison between the two plants because
the manufacturers rated the 12 X 42 drum, at St. Marys, at 27,000 Ibs/hr
evaporative capacity on 78% moisture green chops, whereas the 12 X 36
drum at Topeka was rated at 30,000 Ibs/hr evaporative capacity on 75%
moisture green chop.
The manufacturers estimated that the St. Marys system was recycling
40% of the primary cyclone outlet gases back to the furnace, primary
cyclone, and drum.
54
-------
TABLE 9
THOMPSON PERFORMANCE DATA
Ul
in
Thompson Recycle System
Scale: Full
Approx. Water Usage.
Feeder 6 gpm
Venturl ° 2 gpm
Pellets Green Chops
Prod. Rate Protein Moisture
Test (tons/hr) tt) (I)
1 3.45 17.0 73.0
3 4.63 16.5 73.0
5 4.48 - 76 0
6 4.43 18.5 76.6
B 4.18 18.6 76.0
9 3.85 19.3 77 7
10 3.65 21.6 78.5
ll£/ 4.25 21 4 76.2
12 2.76 20.0 74.9
Water
Usage (Bom) Semolina
Feeder Venturl Location
/ n Inlet (1)
-' ° Stack (2)
0 Inlet (1)
Stack (2)
Recycle to
Primary
Cyclone (6)
4.8 0 Drum (7)
Furnace (8)
0 0 Inlet (1)
Stack (2)
12.0 1 9/ Inlet (1)
Stack (2)
24 2<1./ Inlet (1)
Stack (2)
54 i&l Inlet (1)
Stack (2)
Inlet (1)
0 ° Stack (2)
Stack (2)
. , Recycle to
5 8 2i' o j
J'" *~ Primary
Cyclone (6)
Furnace (B)
Flow Rate
(acfm)
44,500
32,000
43,800
29,100
3,150
2.620
6,470
42,800
32,500
41.900
36,400
40,800
35.700
41,100
36.200
61,200
42,200
34,800
2,700
5,000
Partlc. Loading
(gr/acf)
0.136
0.114
0.328
0.470t/
0.681
0.448
0.314
0.278
0.299
0.211
0.157
0.166
0.082
0.101
0.068
0.148
0.137
0.066
0.135
0.052
Plant. Topeka
Source Operation: Total
Dryer Capacity - 30,000 Ib
Partlc. Rate Z Partic.
(Ib/hr) Recycled
S:5
123.0 4 ,
18.4
10.1
18.8
102-° 18.4
83.2
75-7 35.4
48.9
55.6 54 ,
25.1
35.8 41 l
21.1
52.4
49.3 °
19.5
21.4
3.1
2.2
a/ Without recycle.
b/ Average particle iize = 7 1 u
c/ Volume of water not measured.
d/ Volume estimated.
-------
100.0
WEIGHT % GREATER THAN STATED SIZE
sa si
I
I
C£
LU
b
Test 13(Loc. 5)
;[ Thompson Recycle System
Scale: Full
Plant: Topeka
Source Operation: Total
WEIGHT % LESS THAN STATED SIZE
Figure 22 - Particle Size Distributions - Thompson Recycle System
56
-------
TABLE 10
THOMPSON PERFORMANCE DATA
in
Thompson Recycle System
Plant: St. Marys
Scale:
Test
120
121
122
Pull
Pellets
Prod. Rate
(tons/hr)
3.90
3.90
4.34
Protein Chops Moisture %
(%) Green
21.5 76
20. 7l/ 78.8
20. 7l/ 76.0
Dry
4.2
6.0
5.5.
Source
Operation
D+HMfPM
D+-HM+PM
D4-HM+PM
PWR
Ibs/hr
30,000
34,100
33,600
Evap.
Rate
Ibs/hr
22,400
26,400
25,100
Dryer Capacity = 27,01
@ 78% 1
Partic. Rate Visual
(Ib/hr) Opacity
14.7
17.1
24.5
<10
<=10
I/ Protein content of only one sample taken, during these two tests, by the plant operating personnel. The
pellets which were loaded "on stream" into a rail car during the day of these two tests were reported to
contain 17.2% protein.
-------
Hell Recycle System
Table 11 presents a summary of performance data for the Hell recycle
system at the Grand Island plant. Testing was done with and without
recycle to determine the effect of the recycle system. Without
recycle, emission rates and particulate loadings in the primary
cyclone effluent were high compared to average values measured (at
other plants) in 1971, although the average particle size was fairly
typical. Figure 23 presents the particle size distributions that
were obtained at the outlet and recycle locations.
With the recycle system in operation, the particulate loading and
average particle size at the cyclone outlet were substantially reduced.
However, controlled dryer emissions exceeded the Bay Area limitation
in two out of three cases. The plan was operating between 80% and
90% of capacity.
The test results show the Hell skimmer arrangement to be effective in
concentrating the larger particles in the recycle stream. The grain
loading in the recycle gases was greater by a factor of 1.2 - 3.9
than the grain loading in the outlet gases. This is supported in
test 9 by an average particle size of 7.7 microns in the recycle gases
as compared to an average particle size of less than one micron in the
outlet gases.
There was evidence of considerable buildup of particulate matter on
the interior walls of the extension duct. A cake of 4-5 in. of
deposited material was removed from the bottom interior of the duct
prior to the initial testing. According to a Heil representative,
this material had accumulated during about one month with the recycle
system in operation. The fact that the particulate matter was
preferentially deposited on the bottom of the extension duct indicated
that particulate was settling from the effluent stream; the observed
flow velocity with recycle averaged only about 1,800 ft/min.
Table 12 contains the 1973 test data obtained on the Heil recycle
system at Dundee, Kansas. Tests were not conducted to determine
the emission reduction caused by partial recycling of the primary
cyclone gases back to the furnace. However, the total effect of
recycle, plant design, and operating procedures was a plant with
emissions at 1/3 the allowable levels of the Bay Area Standard.
The tests were run at 84% to 123% of rated capacity based on 34,000
Ib/hr evaporative capacity at 75% moisture in the green chops.
A comparison of test 136 with tests 137 through 139 shows the
tremendous effect of dry chops moisture content on visual opacity.
At 7% moisture in the dry chops, visual opacity was 35-40%,
whereas at 11% dry chops moisture, visual opacity was less than
10%. The effect on total particulate emissions was not appreciable.
The Dundee plant was equipped with a bag filter for control of the
hammermill system emissions.
58
-------
Hell Recycle System
Scale:
Test
1
2*'
3
/,/
5*'
7
9
Full
Pellets
Prod. Rate Protein
(tons/hr) (%)
5.35 19.1
4.83 19.0
4.83 18.8
5.49
5.93
5.22
5.65
Green Chops
Moisture
Ck)
74.0
74.7
74.7
74.2
74.0
73.1
74.6
a/ Without recycle.
b/ Average particle size - 3.0 u.
c/ Average particle size < 1 u.
ij Average particle size = 7.7 11.
TABLE 11
HEIL PERFORMANCE DATA
Flow Rate
(acfra)
31,000
54,800
40,200
46,600
48,700
36,400
36,100
Outlet
Partic. Loading
(Kr/acf)
0.105
0.122
0.138
0.283
0.304^
0.101
Recycle
Partic. Rate
(Ib/hr)
28.0
57.4
47.7
113.2
127.1
31.6
35.7
Flow Rate
(acfm)
-
21,500
_
_
21.000
21,000
Partic. Loading
(Rr/acf^
-
O.S43
_
.
0.125
0.218^
Partic. Rate
(Ib/hr)
-
99.9
.
.
22.6
39.2
Plant- Grand Island
Source Operation: Dryer
Dryer Capacity = 34,000 Ib/hr
7. Partic.
Recycled
68.1
41.7
52.3
-------
100.0
MM Ml Ml
WEIGHT % GREATER THAN STATED SIZE
! i 3 j ! -* »1
10.0
1
Q
LLJ
u
m
Test 9 (Outlet)
i{ . . II '-HI I ..4. ,. !
Test 9 (Recycle)-s
5 (Outlet, Without Recycle)£
Heil Recycle System
Scale: Full
Plant: Grand Island
Source Operation: Dryer J
WEIGHT % LESS THAN STATED SIZE
Figure 23 - Particle Size Distributions - Heil Recycle System
60
-------
TABLE 12
HEIL PERFORMANCE DATA
Hell Recycle System
Scale:
Full
Pellets
Test
136
137
138
139
140
Prod. Rate
Tons/hr
5.62
6.50i/
6.53
4.45
4.75
Protein
(1%)
14.3
18.3
15.7
13.85
14.5
Chops
Green
63.0
72.2
72.2
70.0
71.5
Moisture %
Dry
7.0
11.0
11.0
11.5
9.6
Source
Operation
D
D
D
D
D
PWR
Ibs/hr
27,900
41,600
42,700
26,800
30,900
Evap.
Rate
Ibs/hr
16,800
28,600
29,300
17,700
21,200
Plant: Dundee
Dryer Capacity
Partic. Rate
(Ib/hr)
12.8
6.65
10.8
6.17
9.28
= 34,000 :
Visual
Opacity
38
< 10
<10
0
I/ Estimated
-------
Summary of Recycle
With reference to the recycle systems, the maximum efficiency of
collection and removal of particulate matter would correspond to the
situation in which all of the recycled particulate matter is elimin-
ated by incineration or by discharge from the bottom of the primary
cyclone collector. In this case, the particulate removal efficiency
of the recycle system would be equal to the percent of the particulate
(in the primary cyclone effluent) which is recycled. These percentages
are given in the last columns of Tables 9 and 11.
The total effectiveness of a recycle system in reducing plant emissions
may be greater than the particulate removal efficiency, because of
changes in the drying gases which could reduce the amount of particulate
emissions created in the drum, and should be determined as follows:
% Emission Reduction = (emission w/o recycle-emlssion w/recycle)^ 2.00
emissions w/o recycle
Partial recycling of the primary cyclone gases back to the furnace
has several attractive potential advantages:
1. The substitution of steam for air appears to reduce
the front end temperature enough to make a substantial
improvement in the "smoke" problem, i.e., less smoke
appears to be created with an effectively designed
recycle system.
2. The particulate emissions from the primary cyclone are
apparently reduced at least proportional to the amount
of recycle.
3. There may be a fuel savings depending upon design of
system and rate of production.
4. If a scrubber or other control device must be added,
the net amount of gas to be treated is reduced by the
amount of recycle.
A plant with recycle seems to require a little more operator atten-
tion and it is essential to have the recycle line adequately
insulated to prevent condensation.
62
-------
Effect of Recycle on Product Quality
During some of the tests on the Hell recycle system, samples of raw and
finished product were collected and subsequently analyzed to determine
the effect of recycle on carotene and xanthophyll stability.
Fresh alfalfa was collected from the dehydrator feeder by combining
several grab samples for each condition. The fresh chops were placed
in a plastic bag, quickly frozen with dry ice, returned to the laboratory
and freeze-dried. Corresponding dehydrated meal samples were collected.
After drying and grinding the meals were analyzed for carotene and
nonepoxide xanthophyll (NEX), protein, fiber and meal moisture. A
single sample of each component was collected for experiment 3-0 and
similarly analyzed. The average analyses values for each experiment
are presented in Table 13.
These data indicate that "recycling" had no effect on carotene or
xanthophyll stability under the conditions of these tests. This is
probably because partial recycling of the gases did not reduce the
oxygen content sufficiently to prevent nutrient losses during dehydra-
tion.
The "recycle" meal moistures were slightly higher (6.3 - 10.4%), but
not sufficiently different fro* those of the regularly dehydrated
alfalfa to influence xanthophyll retention. The moisture content
of the alfalfa meal dried under "regular" conditions varied from
5.1 to 7.4%, which is a fairly good range for preventing xanthophyll
losses which occur from overheating and overdrying.
Abilene Plant Modifications
Table 14 depicts the test results obtained on the Abilene plant. Tests
101 through 103 were on extremely low quality material, at high produc-
tion rates and the particulate emissions exceeded allowable limits.
Tests 114-116 were with varying degrees of recycle and indicate a
reduction in emissions as the amount of gases recycled to the furnace
is increased. However, this is contradicted by test 119 which resulted
in a higher level of emissions than test 116 even though both were at the
same level of recycle while test 119 was at a lower production rate and
higher dry chops moisture. Thus, the lower level of emissions during
test 116 may be the result of water spray into the gases just before they
entered the silo.
The plant was within allowable emission levels during tests 141-145 since the
SUB of allowable emissions for each process (drying, grinding, and
pelleting) is greater than the total emissions from the silo.
Although tests 117 and 118 were not run simultaneously with any of the other
tests, it is reasonable to speculate that the particulate contributions
from the hammermill and pellet mill systems were rather high during the
other tests. Also, since the effluent from the hammermill and pellet mill
systems discharged into the silo along with the effluent from the primary
cyclone, this must have confounded the other test results making it
impossible to determine the level of particulate emissions being emitted
63
-------
TABLE 13
EFFECT OF EXHAUST GAS RECYLING ON CAROTENE AND XANTHOPHYLL STABILITY
NEX Meal
Expt. Sample Carotene Xanthophyll Protein Fat Fiber Moisture
mg/lbc mg/lbc % % % %
1-0 Recycle (8/31)a
Freeze-dried 116.1 144.2 19.2
Dehy meal 99.2 123.5 20.3 4.83 25.6 7.7
2-0 Reg. Dehy (9/1)a
Freeze-dried 105.3 140.8 18.7
Dehy meal 105.6 128.5 19.5 4.58 28.9 6.1
3-0 Recycle (9/l)b
Freeze-dried 82.8 110.5 18.0
Dehy meal 88.0 109.8 17.8 4.54 32.2 8.9
aValues for Expt. 1-0 and 2-0 are the av. of three samples, anal, in duplicate.
bValue for Expt. 3-0 is one sample, anal, in duplicate.
cAnalyses are on a moisture-free basis.
-------
TABLE 14
ABILENE PERFORMANCE DATA
Dryer Capacity = 22,000 Ibs/hr
Ul
Pellets
Test
101
102
103
114s/
115k/
116£/
117
118
119£/
14ll/
142
143
144
145
Prod. Rate
(tons/hr)
4.09
4.08
3.78
4.03
4.43
5.07
4.07
4.60
3.56
4.60
3.82
5.10
3.35
2.40
Protein
13.4
11.4
12.1
17.2
16.8
17.3
16.9
14.5
16.8
13.8
19.4
19.1
Chops Moisture %
Green
71.6
64.1
63.1
68.7
67.1
63.5
69.5
75.5
75.8
71.5
77.0
80.3
Dry
12.8
9.0
5.6
11.8
7.9
8.1
11.8
13.5
10.6
12.9
10.3
8.2
Source
Operation
D+HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
HM+PM
HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
D+HM+PM
PWR
Ibs/hr
26,400
20,900
18,700
24,200
25,600
26,500
21,500
34,400
28,900
33,200
26,400
22,700
Evap.
Rate
Ibs/hr
17,800
12,600
11,400
15,600
16,400
16,000
14,000
24,700
21,100
22,300
19,600
17,800
Partic. Rate
(Ib/hr)
66.8
72.5
73.5
51.0
48.2
31.7
48.4
36.9
46.0
28.5
30.6
34.9
31.2
32.9
Visual
Opacity
25
26
48
a_/ Recycling 11% of the primary cyclone effluent back to the furnace
b_/ Recycling 15% of the primary cyclone effluent back to the furnace
£/ Recycling 30% of the primary cyclone effluent back to the furnace
d/ Test only 73.4% isokimetric
-------
from the basic dehydration process. The silo may have collected
some of the paxticulate matter emanating from the hammermill and
pellet mill systems. However, the emission rate from the silo would
undoubtedly have been much less had only the dryer emissions been
discharged into the silo.
Neodesha and Berthoud Plant Modifications
The Neodesha test results, Table 15, are fairly encouraging, even
though the plant was not in compliance during most of the tests be-
cause of the conditions under which the tests were conducted. It was
impossible to run the plant under steady state conditions because
of scarce hay and breakdowns with the harvesting equipment. The hay
also contained a high amount of gases which contributes to excessive
particulate emissions.
The Berthoud plant was not in compliance during its tests, Table 15,
with the most apparent reason being the lower dry chops moisture
content as compared to the Lawrence plant which operated in compliance
even without the wet scrubber in operation, Table 7.
Comparison With The Bay Area Standard
Figures 24, 25, 26, and 27 present plots of the percent of allowable
particulate emissions versus the percent of production capacity for
each of the full-scale control devices/systems that were tested. The
allowable emissions were determined according to the Bay Area emission
standard. The production capacity was calculated on the basis of 75%
green chops moisture, 8% dry chops moisture and 7% pellet moisture
and on the evaporative capacity of the dryer. Figure 25 shows two plots,
connected by a horizontal line, for each of the 1973 Thompson tests.
The plots indicating the higher percent of production capacity were
derived from a plant evaporative capacity based on 78% moisture green
chops. The Thompson plots indicating lower percentage of production
capacity and the plots for the other plants were derived from evapora-
tive capacities based on 75% moisture green chops.
It should be noted that those plants with more than one source opera-
tion (shown on previous tables) have higher allowable emissions for a
given process-weight-rate than the plants with only the one source
operation, the dryer. This is because, for the multi-source
operations, the allowable emissions for each process (drying, grinding,
and pelleting) are totaled to give one allowable emission from the
common discharge point for all the process; e.g., a plant with a
process-weight-rate of 20,000 Ibs/hr through the dryer would be allowed
19.2 Ib. of emissions per hour from the primary cyclone if the primary
cyclone is handling only the dryer discharge; whereas, another plant
with the same process-weight-rate through the dryer would be allowed
29.6 Ibs. of emissions per hour from the primary cyclone If it was
processing 8,000 Ibs. per hour through the hammermill and the hammer-
mill system effluent was also being discharged through the primary cyclone.
66
-------
TABLE 15
NEODESHA PERFORMANCE DATA
Dryer Capacity = 20,000 Ibs/hr
a*
Pellets
Test
user
124
125£/
IMS/
127S/
12 a*/
129
Prod. Rate
(tons/hr)
2.97
2.97
3.20
3.10
2.77
2.97
2.70
Protein
17.3
18.5
17.7
17.6
14.8
15.8
16.8
ji/ Recycling approximately 10%
Chops Moisture %
Green
79.1
76.8
73.8
70.6
76.4
71.4
67.0
of the
Dry
14.9
10.0
15.5
13.6
14.1
13.5
Source
Operation
D+HM
D+HM
D+HM
D+HM
D+HM
D+HM
15.5 D+HM
primary cyclone effluent
PWR
Ibs/hr
26,600
24,300
23,000
19,800
21,900
19,300
Evap.
Rate
Ibs/hr
20,100
18,000
15,900
13,100
15,900
12,900
15,300 9,300
back to the furnace.
BERTHOUD PERFORMANCE DATA
133
134
135
3.15
2.05
2.80
18.2
19.9
19.7
73.7
82.0
80.5
10.3
8.4
10.7
D+HM
D+HM
D+HM
22,300
21,300
26,500
15,800
17,100
20,700
Par tic. Rate Visual
(Ib/hr) Opacity
28.2
44.0
28.9 34
35.3
43.4 32
42.2
23.7
Dryer Capacity - 20,000
46.5
31.7
45.8
-------
400 r
-
_
-
-
V
o
3
a
A
PLANT
Lexington
Oxford
Topeka
Grand Island
Rozel
CONTROL
EQUIPMENT
Air Cond
Koch
Thompson
Heil
Koch
300 -
1/1
z
o.
-Jl
Closed Symbol = Without Control
Open Symbol = With Control
20
!*~2-Test Number
I
PERCENT OF ALLOWA
_ NJ
8 §
4
1
3
T
Bay Area Emission Standard
\
1 1 1
P
^
f
^
r
\
r
m 6
r
3
8D0
1 ' n D 5 i |
' , i * /
T *
7 0 A
i
40 60 80
PERCENT OF PRODUCTION CAPACITY
100
120
Figure 24 Control Effectiveness vs Bay Area Emission Standard
-------
240
A Dundee (D)
O St. Mary's (D+H+P)
220
Open Symbol = with Control
200
180
160
g
17?
UJ
CO
Q120
Bay Area Standard
100
Z
UJ
UJ
a.
80
60
136
A
0122
40
20
140
139
137
50 60 70 80 90 100 110 120 130
PERCENT OF PRODUCTION CAPACITY
FIGURE 25
-------
240
A Oxford (D)
B Lawrence (D+H)
220
Open Symbol =with Control
200
180
Z
o
tri
CO
§140
149
0120
130
O Bay Area Standard
100
Z
LU
LU
a.
80
1110 B109
111
60
A105 147 D146
20
0132
40
0112
0113
50 60 70 80 90 100 110 120 130
PERCENT OF PRODUCTION CAPACITY
FIGURE 26
IQ
-------
*~ 9 Abilene (D+H+P)
* Neodesho (D+H)
D Berthoud (D+H)
220
Open Symbol = with Control
200
0103
180
0102
160
. 0133
Z
O A127 0101
r^TD135
140 124
< A126
§120- Q D134 0114
O Bay Area Standard
>_100 X
^ °145 A123
a.
80
0144
0142 OO143
116
60
0141
40
20
\ I I I I
50 60 70 80 90 100 110 120 130
PERCENT OF PRODUCTION CAPACITY
FIGURE 27
71
-------
Visual Opacity
Visual opacity readings were made at some of the plants, by qualified
State Control Department personnel, during the 1972 and 1973 source
tests. Table 16 shows these visual opacity readings and the
corresponding measured grain loadings and pounds of emissions per hour.
The visual opacity is expressed as the average of all readings taken
during a source test.
TABLE 16
VISUAL OPACITY READINGS
Farticulate Emissions
Test
9-2
10-2
11-2
121
122
125
127
131
132
136
137
138
141
142
143
Visual Opacity %
39
17
45
< 10
< 10
34
32
«= 10
< 10
38
< 10
< 10
25
26
48
gr/acf
.082
.068
.137
.061
.093
.089
.135
.049
.045
.066
.028
.046
.073
.079
.092
Ibs/hr
25.1
21.1
49.3
17.1
24.5
28.9
43.4
13.3
12.2
12.8
6.65
10.8
28.5
30.6
34.9
72
-------
SECTION VIII
EFFECT OF PROCESS CONDITIONS
The results of ADA's 1971 study indicated a strong dependence of
dryer-generated particulate emissions on input hay quality and
dryer operating conditions. This section of the report quantifies
that dependence through single-plant and multiplane correlations.
The multiplant correlation includes data from the 1971 study.
This section also presents a comparison of cyclone collection
efficiency and a tabulation of emission factors derived from the
results of the 1971 and 1972 studies. Factors are given for
the grinding and pelleting operations as well as the drying opera-
tion.
Single-Plant Correlation (Topeka)
Analysis of the performance data for a single plant indicated that
the particulate emission rate from the primary cyclone collector
was strongly dependent on the dry chops (or pellet) production
rate and on hay quality (pellet protein). The production rate is
a measure of the rate at which dry matter passes through the dryer.
Multiple regression techniques were used to correlate the data
from the Topeka plant, which exhibited the greatest ranges of
variation in process parameters. A high degree of correlation
(multiple correlation coefficient = 0.98) was obtained with the
following equation:
(Prod. Rate)3-0
Emission Rate = 710 X
(% Protein)2'2
where the emission rate is expressed in Ib/hr nd the production
rate in ton/hr.
This equation states that emissions nearly double at a constant produc-
tion rate when the protein is decreased from 21% to 16%. Moreover,
there is a cubic dependence on production rate; for example, if the
production rate is increased from 75% of capacity to 100% of capacity,
the rate of emissions is more than doubled. Although the number of
data points used in developing this correlation equation was not
large, the high degree of correlation indicates that this equation
should be reliable in predicting the sensitivity of emissions to these
two process parameters for this plant.
Since low quality hay and high production rate normally occur together,
there is a very large difference in emissions possible between operat-
ing conditions during the high productivity period (midsummer) and the
low productivity period (early and late summer).
73
-------
Multiplant Correlation (Fine Particle Emissions)
Analysis of multiplant data (including data from the 1971 study)
by multiple regression techniques also substantiated the dependence
of the particulate emission rate on process parameters. However,
the degree of correlation was sharply reduced due to the
dissimilarities in plant operating characteristics, i.e., differences
in dryer size and performance and in primary cyclone collection
efficiency. The differences in dryer size were compensated for by
introducing percent of production capacity into the correlation
instead of the absolute production rate, and separating the dryers
into two size categories. The differences in cyclone collection
efficiency were compensated for by considering only fine particle
(less than 5 u) emissions, which are not efficiently collected by
any of the cyclones.
For the dryers in the smaller capacity category (15,000 to 20,000
Ib/hr of evaporation), a high degree of correlation (multiple
correlation coefficient = 0.91) between fine particle emissions and
process conditions was obtained with the following equation:
(Z of Prod. Cap.)1-45
Emission Rate3-' =306,OOOX
(% Protein) 5'67
A graph of the above equation is presented in Figure 28.
For the dryers in the larger capacity category (30,000 to 34,000 Ib/hr
of evaporation), only a moderate degree of correlation (multiple
correlation coefficient = 0.71) could be obtained, and the
reliability of the correlation equation, as a predictive tool, was
correspondingly reduced.
In both the single plant and multiplant correlations between dryer-
generated particulate emissions and process conditions, emissions
(total and fine particle) were found to increase with increasing
production rate and with decreasing hay quality.
for particles <5 microns
74
-------
LU
35
30
25
20
O 15
LO
to
LU
£ 10
u
5
° 306.000
20
(% Protein)
5'67
40 60
PERCENT OF PRODUCTION CAPACITY
80
Figure 28 - Calculated Fine Particle Qnissions (<5p) -- (Smaller Dryers)
-------
Cyclone Collection Efficiency
To compare the relative collection efficiencies of the primary
collectors involved in this study and in the 1971 study, the
assumption was made that the size distribution of the total
particulates entering a primary collector did not vary significant-
ly from one plant to another. Based on this assumption, high
efficiency is indicated by a small fraction of large particles
(> 5 ;u) in the collector effluent. The results of this comparison
are shown in Table 17; as expected, the standard conical cyclones
are indicated to be the most efficient.
This, plus the grouping of the collectors by type, supports the
assumption on which Table 17 was based.
TABLE 17
COMPARATIVE EFFICIENCIES
:F PRIMARY "V.-LQHT COLLECTORS
WT % of
Large Particles
Collector Type (> 5 u) in Emissions
Conical 25
Conical 40
Conical 45
Conical 45
Flat-bottom 50
Flat-bottom 50
Conical 55
Flat-bottom 60
Cylindrical 80
76
-------
Emission Factors
Table 18 presents a summary of uncontrolled emission factors
for the drying, grinding and pelleting operation, and combinations
thereof. These factors were derived from results of the 1971 tests
of Plants A-D (1) and the 1972 tests of plants E-I. In the latter
case, the uncontrolled emissions are the values at the inlets to
control equipment or with recycle flows shut off.
As shown in Table 18 , the total plant emissions average less
than 20 Ib/ton of pellets (or meal). This is significantly less
than the value of 60 Ib/ton of meal reported by the U.S.
Environmental Protection Agency (7) and commonly used by the
states in estimating emissions from alfalfa dehydrating mills.
Table 18 shows the emissions from each source on plants which
have separate outlets to atmosphere from the dryer, hammermill
and pellet mill cyclone collectors. Plant D has one outlet for
both the dryer and hammermill cyclones. Plants A, B and E
return the hammermill and pellet mill cyclone effluent to the
primary "dryer" cyclone and thus have only one emission source
for the total plant. However, test A-102 is listed as dryer +
hammermill because the pelleting system was shut down during this
test and the one emission source contained only the effluent from
the dryer and hammermill.
77
-------
TABLE 18
UNCONTROLLED EMISSION FACTORS
Source Operation
Average Emission Factor
(Ib/ton of green chops)
(Ib/ton of pellets)
Dryer
C-307
F-2,4,5
G-1,3
H-1,2,4,6
1-2,3,4
6.55
4.98
1.42
1.71
3.12
25.27
17.98
6.78
5.82
15.27
oo
Dryer + Hammermill
Hammermill
Mean:
St.
A-102
D-405,406,408,410
C-305
3.55
2.19
Mean:
St. Dev:b/
Mean:
St.
14.22
8.11
18.58
9.98
14.28
6.08
2.47
Pellet Mill
C-302
D-402,403
Mean:
St. Dev:t/
2.66
1.33
2.00
0.94
Total
A-101
B-202,204,209,211
E-ll
18.22
21.84
12.33
&/ Letter designations for plants are given in Appendix D.
b/ Standard deviation.
Mean:
St. Dev:£/
17.46
4.80
-------
SECTION IX
CONTROL COSTS - MODEL PROBLEM
For the evaluation of the comparative costs of the control of
dryer-generated particulate emissions, each equipment manufacturer
was asked to submit performance and cost data applicable to the
control of emissions from a "model" dryer/primary cyclone. The
effluent properties which were specified for the model problems are
representative of a typical alfalfa dehydrating plant but do not
necessarily reflect the severest air pollution conditions which might
exist. Make-up water properties and wastewater impurity limits
were also given.
Table 19 presents the model problem data form. The requested
performance data included the gas pressure drop and water usage
for a control device/system which would meet or exceed the
specified collection efficiency of about 60%. The requested
cost data included costs of purchasing, installing and operating
the control device/system and costs for water treatment.
Table 20 gives a summary of cost data submitted by each manufac-
turer. As indicated in the table, varying degrees of cost breakdown
were specified in the completed forms.-
The range of variation in the cost figures suggests that there were
significant differences in cost estimating techniques and/or in
cost items included in the totals. For example, it is difficult
to imagine that the variation in ductwork costs (Item A-3) for a
wet scrubber could be as large as shown. A second example would be
the inclusion of freight costs; only Fisher-Klosterman
specifically mentioned this item. Also it should be noted that
the Koch figures are based on the installation of the scrubber on
top of the primary cyclone and on the use of a sand-lined settling
pond for disposal of scrubber liquid.
The wide variations in installation costs presented by the companies
is partly because the model problem did not delineate specific
installation parameters. However, it is difficult to imagine that a
wet scrubber could be installed on any dehydrating plant for only
$1,100 as shown by one company. Since the companies obviously used
different assumptions in arriving at their installation costs,
these figures cannot be used to compare the installation costs of one
wet scrubber to another. These figures simply show that wet scrubber
installation costs may range from $5,000 - $20,000.
In general the manufacturers will "guarantee" performance of emission
control equipment if such equipment and/or the plant are operated
in accordance with the manufacturer's instructions. The "guarantees"
are also predicated on the uncontrolled plant being generally no more
difficult to control than the model plant. However, the extent of
the manufacturer's liability for equipment performance is usually not
stated and therefore requires clarification.
79
-------
TABLE 19
MODEL PROBLEM DATA FORM
PERFORMANCE AND COST DATA
- MODEL CONTROL PROBLEM -
Company
Information submitted by
Signature Date
Phone No,
I. Description of Model Control Problem
A. Effluent Properties (Primary Cyclone)
1. Flow rate = 30,000 acfm
2. Temperature - 225°F
3. Moisture content = 37% by vol.
4. Particulate loading = 0.20 grains/acf
5. Particulate size distribution:
85% finer than 20 u
67% finer than 10 u
51% finer than 5 P
40% finer than 2 n
33% finer than 1 u
6. Uncontrolled emission rate = 51 Ib/hr
7. Maximum allowable emission rate = 20 Ib/hr
B. Water Properties
1. City water available at 80°F, 250 mg/1 total dissolved
solids, virtually free of suspended solids and BOD.
2. Wastewater impurity limits are < 500 mg/1 IDS, < 30 mg/1 SS,
< 30 mg/1 BOD, < 50 mg/1 COD, 6.5 < pH < 8.5.
80
-------
TABLE 19 (Continued)
II. Performance Data
A. Pressure drop required = in. of
B. Water required:
Recirculation rate = GPM at psi AP
Make-up rate = GPM
C. Expected particulate collection efficiency = 7, by weight
Would you guarantee, based on the specified operating conditions,
that emissions will not exceed the 20 Ib/hr limit?
Yes No
Would you be willing to bear the cost of any performance test
by a mutually acceptable testing organization?
Yes No
Comments:
III. Cost Data
A. Estimated equipment cost - including collector, ductwork, fans,
pumps, water treatment facilities, and all appurtenances
= $ .
B. Estimated installation cost (for all of the above) = $ .
C. Estimated total installed cost (A + B) = $ .
D. Estimated annual operating cost (water and electricity), assuming
3,000 operating hour per year = $ .
Comments:
81
-------
TABLE 20
MODEL PROBLEM COST DATA-/
Manufacturer
F/K
2. Fan
3. Ductwork
4. Other
Total
B. Water Treatment
1. Clarifier
2. Pump
3. Other
Total
C. Total
II. Installation
A. Materials
B. Labor
C. Total
III. Operation (3,000 hr)
A. Power
(Pressure Drop
H20) 6 in.
B. Water
(Recirc.) 122
(Make-up)8pm 2
C. Labor
D. Total
APPCOR
ACC
Koch
Hell
I. Equipment
A. Particulate
Collection
1. Collector 9,850 7,500
5,600
8,000
n.r.
n.r.
n.r.
$ 6.500 $ 5.500
$2.000
n.r.
7 in.
60
3.7
$ 2.500 $ 2,090
$ 450
3 in. 3 in.
$ 50
405 125
7.2 12
$ 500
$ 800 $ 1.200
n.r.
Thompson
n.r.
12,350 1,000
$22.200 $14.100 $13.300 $8.000 $17.800 $17.000
n.r.
1,000
1,000
n.r.
$28.700 $19.600 $13.300 $10.000 $17.800 $17.000
$17.600 $15.000 $ 1.100 $ 5.000 $ 1.600 $ 3.000
0-8
($ 2.300)^/($ 6.000)^
aj n.r. - item not required
by Fuel savings figures as provided by the manufacturers.
82
-------
SECTION X
SAMPLING LOG, PERFORMANCE AND PROCESS DATA,
AND PLANT EQUIPMENT SPECIFICATIONS
Table 21 presents the sampling log, i.e., the time during which
each test was conducted, the dehydrating plants where tests were
conducted, the control devices/systems that were tested and the
sampling locations. The table also indicates the source opera-
tions, i.e., drying (D) and hammexmilling (HM), which contribute
particulate matter to the effluent from the primary cyclone.
Test numbers ending in "A" denote particle sizing tests
conducted with the Andersen in-stack impactor. In all tables,
"I" refers to an inlet sample, and "0" refers to an outlet sample.
Tables 22-29 present performance data for each of the particulate
emission control devices/systems which were tested and process data
for each of the plants where the control equipment was installed.
Table 22 lists the factors used in calculation of the quantities
listed in Tables 23 through 36.
Table 37 lists information on the rotary drum dryers at the alfalfa
dehydrating plants which were utilized in the studies reported
herein.
83
-------
Plant/Source Operation
Neodesha/D+HM
Control Device/
System
Fisher-Klosterman
(pilot scale add-on)
Topeka/Total
Thompson System
(full scale)
Neodesha/D+HM
Grand Island/Dryer
Oxford/Dryer
Fisher-Klosterman
(pilot scale add-on)
Hell System
(full scale)
Koch
(full scale add-on)
TABLE 21
SAMPLING LOG
Sampling Location
(inlet
[Outlet
Inlet
{Inlet
Dutlet
Inlet
(inlet
[Stack
flnlet
[Stack
flnlet
[Stack
Stack
Recycle to Primary Cyclone
Recycle to Drum
Recycle to Furnace
Inlet
[Stack
Stack
flnlet
{Stack
fin let
(Outlet
flnlet
(Outlet
Outlet
Outlets/
Outlet
^Recycle to Furnace
rinlet
Jputlet
"inlet
[Outlet
'inlet
^Outlet
Test
Number
1-1
1-0
2-IA
3-1
3-0
4-IA
1-1
1-2
2-IA
2-2A
3-1
3-2
4- 2 A
5-6
5-7
5-8
6-1
6-2
7-2A
8-1
8-2
5-IA
5-OA
6-1
6-0
1-0
2-0
3-0
3-R
1-1
1-0
2-IA
2-OA
3-1
3-0
Time
Date
7-26-72
7-26-72
7-26-72
7-27-72
7-27-72
7-27-72
8-08-72
8-08-72
8-09-72
8-09-72
8-09-72
8-09-72
8-10-72
8-10-72
8-10-72
8-10-72
8-11-72
8-11-72
8-11-72
8-11-72
8-11-72
8-23-72
8-23-72
8-23-72
8-23-72
8-31-72
9-01-72
9-01-72
9-01-72
9-07-72
9-07-72
9-07-72
9-07-72
9-08-72
9-08-72
Start
2:22 p.m.
2:25 p.m.
5:21 p.m.
2.24 p.m.
2:25 p.m.
3:26 p.m.
4:40 p.m.
4:40 p.m.
1:13 p.m.
1:15 p.m.
3:59 p.m.
3:59 p.m.
10:30 a.m.
1:05 p.m.
1:05 p.m.
1-08 p.m.
9.02 a.m.
9:02 a.m.
1:24 p.m.
2:06 p.m.
2:06 p.m.
12:55 p.m.
12:52 p.m.
2:30 p.m.
2:30 p.m.
12:31 p.m.
11:34 p.m.
4:43 p.m.
4:50 p.m.
1:45 p.m.
1-15 p.m.
5:37 p.m.
5:48 p.m.
2:17 p.m.
1:35 p.m.
Sampling
Finish Duration (min)
4:18 p.m.
4:21 p.m.
5:41 p.m.
4:36 p.m.
4:38 p.m.
3:55 p.m.
7-17 p.m.
7.13 p.m.
1:19 p.m.
1:21 p.m.
7:39 p.m.
7:11 p.m.
10:30 a.m.
1:56 p.m.
1:41 p.m.
1:44 p.m.
11:25 a.m.
11:25 a.m.
1:25 p.m.
3.54 p.m.
3:54 p.m.
1:20 p.m.
1:17 p.m.
4:00 p.m.
4.00 p.m.
3:25 p.m.
1:28 p.m.
5:19 p.m.
5:18 p.m.
4:06 p.m.
5:11 p.m.
5:52 p.m.
6:18 p.m.
4:35 p.m.
5:15 p.m.
60
60
20
60
60
25
51
72
6
6
52
52
1/4
36
36
36
72
72
1
72
72
25
25
60
60
72
96
36
28
80
180
15
30
80
180
-------
TABLE 21 (Continued)
Plant/Source Operation
Grand Island/Dryer
Control Device/
System
Heil System
(full scale)
Topeka/Total
Thompson System
(full scale)
Neodeaha/D+HM
APPCOR
(pilot scale add-on)
Lexington
Air Conditioning Corp.
(full scale add-on)
Rozel/Dryer
Koch
(full scale add-on)
Sampling Location
Outlet^/
Outlet!/
Outlet
Outlet
Recycle to Furnace
Outlet
Recycle to Furnace
'outlet
^Recycle to Furnace
'inlet
.Stack
'inlet
Stack
Inletfi/
.Stack*/
[Stack
Recycle Co Primary Cyclone
[Recycle to Furnace
Downstream - 2nd Venturl
'inlet
.Outlet
flnlet
[Outlet
Inlet
Cutlet
[inlet
[Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
[inlet
but let
[Inlet
Outlet
Inlet
Outlet
Inlet
Outlet
Inlet £'
Inlet y
Test
Number
4-0
5-0
6-OA
7-0
7-R
8-OA
8-RA
9-0
9-R
9-1
9-2
10-1
10-2
11-1
11-2
12-2
12-6
12-8
13-5A
1A-IA
1A-OA
2A-I
2A-0
3A-I
3A-0
4A-I
4A-0
1-IA
1-OA
2-1
2-0
3-1
3-0
4-1
4-0
1-1
1-0
2-1
2-0
3-IA
3-OA
4-1
4-0
5-IA
6-1
Time
Date
9-12-72
9-13-72
9-13-72
9-14-72
9-14-72
9-14-72
9-14-72
9-15-72
9-15-72
9-19-72
9-19-72
9-20-72
9-20-72
9-20-72
9-20-72
9-21-72
9-21-72
9-21-72
9-21-72
9-27-72
9-27-72
9-27-72
9-27-72
9-28-72
9-28-72
9-28-72
9-28-72
10-04-72
10-04-72
10-04-72
10-04-72
10-05-72
10-05-72
10-05-72
10-05-72
10-11-72
10-11-72
10-11-72
10-11-72
10-12-72
10-12-72
10-12-72
10-12-72
10-13-72
10-13-72
Start
3.25 p.m.
12 50 p.m.
3:30 p m.
1:48 p.m.
1-48 p.m.
5-20 p.m.
5-05 p.m.
11:35 a.m.
11 35 a.m.
2 36 p.m.
2.36 p.m.
11:00 a.m.
11:00 a.m.
3.22 p.m.
3:22 p.m.
1.45 p.m.
1:53 p m.
1 53 p.m.
4 54 p.m.
2 39 p.m.
2 35 p.m
4-23 p.m
4.23 p.m.
10:18 a.m.
10:18 a.m.
1-57 p.m.
1:57 p.m.
3:50 p.m.
3 06 p.m.
5:30 p.m.
5:31 p.m.
10:13 a.m.
10:14 a.m.
1.30 p.m.
1.32 p.m.
1:30 p.m.
1-30 p.m.
5 45 p.m.
5:45 p.m.
10.05 p.m.
9:55 a.m.
2:25 p.m.
2:25 p.m.
11 45 a.m.
3.30 p.m.
Finish
5.06 p.m.
4.36 p.m.
3:39 p.m.
3:56 p.m.
3 56 p.m.
5:29 p.m.
5.13 p.m.
1:08 p.m.
1-03 p.m.
5:36 p.m.
5 36 p.m.
12 51 p.m.
12.51 p.m.
4:59 p.m.
4.59 p.m.
3 36 p.m.
3:28 p.m.
3:30 p.m.
5:00 p.m.
3:06 p.m.
3:10 p.m.
5:50 p.m.
5-50 p.m.
11 28 a.m.
11 28 a.m.
4:26 p.m.
4:26 p m.
4.10 p.m.
3-46 p.m.
7.13 p.m.
7:09 p.m.
11:25 p.m.
11-30 p.m.
3 54 p.m.
3:58 p.m.
3.42 p.m.
3:35 p.m.
7 12 p.m.
7:12 p.m.
10:30 p.m.
10:45 p.m.
3 55 p.m.
3.55 p.m.
12 15 p.m.
4-42 p.m.
3 at;ling
Du-pt-on (min)
72
72
9
72
64
9
8
72
64
72
72
72
72
72
72
72
48
48
7
16
24
48
48
48
48
48
48
20
40
60
64
48
65
48
64
60
60
60
60
24
50
60
60
30
60
b/
Without recycle.
Unit disconnected.
-------
Control Device/
Plant/Source Operation System
Abilene/D+HMfPM
Plant Modifications
TABLE 21 vContinued)
Sampling. Location
Silo
Outlet
HM+PM
Line to Silo
Tesu
Number
101
102
103
114
115
116
117
118
115
141
142
143
144
145
Sampling
Duration
Date
6-15-73
6-18-73
6-18-73
7-10-73
7-11-73
7-11-73
7-12-73
7-12-73
7-13-73
9-4-73
9-4-73
9-6-73
9-7-73
9-7-73
Start
10:
11:
6:
2:
10:
2:
11:
3:
10:
2:
5:
1:
10:
2:
50
00
30
54
52
45
40
00
15
20
28
22
08
04
a
a
P
P
a
P
a
P
P
P
P
P
a
P
.m.
Hi,
.m
.m.
.JE.
nu
.nil
.m.
.m.
.m.
.m.
.m.
.m.
.m.
Finish
12
12
7
4
12
20
30
40
35
26
3:57
12
4
11
4
7
3
11
3
37
14
31
19
22
06
29
28
P
P
P
P
P
P
P
P
P
P
P
P
P
P
. nic
.m.
.m.
.m.
.m.
.m.
.m.
.m.
. m.
.m.
.m.
in*
in*
in.
(min.)
48
48
48
64
64
64
48
48
64
64
64
64
64
64
oo
Oxford/D
Lawrence/D+HM
Dundee/D
Koch
(Full Scale add-on)
None
None
None
Koch (Full Scale)
ii ii ii
None
Koch (Full Scale)
II M II
II II II
II II II
None
Heil System
(Full Scale)
Outlet
Stack to ATM
n ii n
n ii n
Outlet
Stack to ATM
Outlet
Stack to ATM
Outlet
ti
n
n
104
105
109
110
111
112
113
130
131
132
146
147
148
136
137
138
139
140
6-20-73
6-20-73
7-5-73
7-6-73
7-6-73
7-9-73
7-9-73
8-6-73
8-7-73
8-7-73
9-10-73
9-11-73
9-11-73
8-28-73
8-29-73
8-29-73
8-30-73
8-31-73
9
4
2
9
1
10
2
4
12
3
4
3
6
3
11
3
5
1
:55
:20
:00
:47
:56
:45
:45
:13
:45
:15
:20
:12
:06
:12
:39
:39
:45
:42
a
P
P
a
P
a
P
P
P
P
P
P
P
P
P
P
P
P
.m.
in.
.m.
.m.
.m.
.m.
nu
.m.
m«
.m.
.m.
.m.
>m>
. m.
.m.
.m.
. m.
.m.
1
7
4
11
3
12
4
5
2
4
6
4
8
5
1
4
7
3
:00
:30
:00
:21
:23
:06
:18
:42
:07
:35
:30
:57
:15
:19
:15
:15
:23
:15
P
P
P
a
P
P
P
P
P
P
P
P
P
P
P
P
P
P
m.
.m.
. m.
.m.
.m.
.m.
.m.
.m.
. m.
.m.
.m.
.m.
.m.
in*
m.
.m.
.m.
.m.
144
160
72
72
72
72
72
72
72
72
72
72
69
72
72
36
72
72
.8
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
-------
TABLE 21 (Concluded)
Control Device/
Plant/Source Operation System
Neodesha/D+HM
Plant Modifications
St. Marys/DfBM+PM
Thompson
System
Sampling
Sampling Location
Stack to ATM
Outlet
Test
Number
123
124
125
126
127
128
129
120
121
122
Date
7-31-73
7-31-73
8-1-73
8-1-73
8-2-73
8-2-73
8-3-73
7-16-73
7-27-73
7-27-73
Time
Start
11:35 a.m.
4:27 p.m.
3:27 p.m.
6:41 p.m.
11:18 a.m.
3:00 p.m.
10:45 a.m.
2:31 p.m.
12:00 a.m.
3:15 p.m.
Duration
Finish
12:15 p.m.
5:57 p.m.
4:57 p.m.
8:00 p.m.
12:42 p.m.
4:30 p.m.
12:18 p.m.
3:58 p.m.
1:15 p.m.
4:15 p.m.
(min)
72.0
72.0
72.0
72.0
72.0
72.0
72.0
72.0
72.0
72.0
-------
TABLE 22
PERFORMANCE AND PROCESS DATA CALCULATION FACTORS
Value Determined
Units
Factors in Calculation
Dryer
Conditions
Carrier Gas
Conditions
oo
oo
Control
Performance
Process weight rate
Back-end temperature
Excess air
Evaporation rate
Fuel rate
Temperature
Moisture
Average velocity
Plow rate
Partlculate loading
Pellet production rate
Moisture flow
Pressure drop
Water usage rate
Average particle size
Partlculate rate
Collection efficiency
Ib/hr Production rate, green chop moisture, pellet moisture
°F Average dryer outlet temperature
7, Carrier gas composition (dry basis)
Ib/hr Process weight rate, green and dry chop moisture
scfh Meter readings, elapsed time, average meter pressure and temperature
°F Average carrier gas temperature
7. by vol. Moisture condensate, volume of carrier gas sampled
fpm Average velocity head, temperature, pressure, and composition of carrier gas
acfm Average velocity, cross-sectional area
dscfm Flow rate (acfm), temperature, pressure, and composition of carrier gas
gr/acf Parclculate weight, volume of carrier gas sampled
gr/dscf Particulate loading (gr/acf), temperature, pressure, and composition of carrier gas
tons/hr Number of scale dumps, elapsed time
Ib/hr Flow rate, temperature, pressure, and moisture content of carrier gas
in. HjO Carrier gas pressures (inlet and outlet)
gpm Water meter readings, elapsed time
microns Andersen graphs (SOU of partlculate weight less than average size)
Ib/hr Partlculate loading, flow rate of carrier gas
% Partlculate rates (inlet and outlet)
-------
TABLE 23
GO
NO
Fisher-Klosterman
(Neodesha/m-HM)
Process
Weight
Rate
Test
FISHER-KLOSTERMAN/NEODESHA PERFORMANCE AND PROCESS DATA
Test
1-1
1-0
3-1
3-0
6-1
6-0
20,600
22,000
19,300
Temp.
208
164
206
166
182
217
Pellet Prod.
Rate
Test (tons/hr)
1-1
1-0 .
3-1
3-0 .
> 2.85
> 2.62
6-1 1
6-0 r 3-°°
Protein
(7.)
18.5
18.5
~
Moisture
(7. by Vol.)
32.7
31.7
37.3
31.0
24.1
16.4
Moisture
Flow
(Ib/hri
605
430
608
458
361
269
DRYER CONDITIONS
Back -End
Chop Moistures (%) Temp .
Green Dry (°F)
74. 3-/ 10. 8^ 270
77.95/ 11. IS/ 258
71.2 6.9 250
CARRIER GAS CONDITIONS
Avg. Vel. Flow Rate
(fpm) (acfm) 'dscfm)
1,734 859 441
1,205 589 333
1,517 731 362
1,304 639 364
1,366 672 407
1,525 750 491
CONTROL PERFORMANCE
Excess
Air
(7.)
202
182
512
Evaporation
Rate
(lb/hr)
14,600
16,500
13,300
Particulate Loading
(gr/acf) (sr/dscf)
0.
0.
0.
0.
0.
0.
Pressure Water Avg. Particle Particulate
Drop Usage Size
(in. H,0) (gpm) (u)
12 4.4 < L
11.7 3.1 < l
< 1
9.5 3.7
Rate
(Ib/hr)
0.45
0.28
0.53
0.31
0.29
0.12
0617 0.1204
0549 0.0972
0818 0.1699
0561 0.0985
0497 0.0822
0183 0.0280
Collection
Efficiency
(7.)
37.8
41.5
58.6
a/ Laboratory result.
-------
TABLE 24
APPCOR
(Neodesha/W-HM)
APPCOR/NEODESHA PERFORMANCE AND PROCESS DATA
SO
o
Test
2
3
4
Test
2-1
2-0
3-1
3-0
4-1
4-0
Test
2-1
2-0
3-1
3-0
4-1
4-0
Process
Weight
Rate
(Ib/hr)
22,700
16,800
16,100
Temp.
216
154
193
153
218
155
Pellet Prod.
Rate
(tons/hr)
2.75
1.95
2.15
DRYER CONDITIONS
Protein
(7.)
21.8
18.9
19.9
Moisture
("; by Vol.)
32.1
33.1
32.0
32.2
32.9
33.2
Sack-End
Chop Moistures (7.) Temp .
Green
76.8^
77.3a-/
73.8S-/
CARRIER
Avg. Vel.
(fom)
4,410
4,500
3,980
3,960
4,160
4,000
Drv (°F)
6. 52.' 275
4.0S./ 290
5.2S/ 285
GAS CONDITIONS
Flow Rate
(acfm) (dscfm)
2,230 1,160
2,270 1,260
2,010 1,080
2,000 1,120
2,100 1,060
2,020 1,100
Excess
Air
C*>
353
Evaporation
Rate
(Ib/hr)
17,100
12,900
11,600
Particulate Loading
igr/acf) (gr/dscf)
0
0
0
0
0
0
.0943 0.1821
.0389 0.0705
.1209 0.2265
.0385 0.0692
.1732 0.3430
.0606 0.1110
CONTROL PERFORMANCE
Moisture
Flow
(Ib/hr)
1,540
1,740
1,430
1,480
1,470
1,530
Pressure
Drop
(in. H,0)
7.9
7.5
7.5
Water Avg. Particle Parti culate
Usage Size
(Epm) (u)
fib/ 11.0
< 1
6fe/ :
&!>/
Rate
(Ib/hr)
1.80
0.76
2.09
0.66
3.12
1.05
Collection
Efficiency
(%)
57.8
68.4
66.3
a/ Laboratory results.
b/ Estimated value.
-------
TABLE 25
Koch
(Oxford/Dryer)
KOCH/OXFORD PERFORMANCE AND PROCESS DATA
Test
1
3
Test
1-1
1-0
3-1
3-0
Test
1-1
1-0
3-1
3-0
Process
Weight
Rate
(Ib/hr)
19,200
14,500
Temp.
207
160
219
158
Pellet Prod.
Rate
(tons/hr)
2.07
1.49
Protein
(7.)
20.9
18.3
Moisture
(% by Vol.)
44.8
35.1
41.8
33.7
Moisture
Flow
(lb/hr)
18,400
13,600
16,700
13,100
DRYER CONDITIONS
Back- End
Chop Moistures (%) Temp.
Green Dry ( °F)
80. 3§7 8.4S/ 229
81. 027 7.4l/ 232
CARRIER GAS CONDITIONS
Avg. Vel. Flow Rate
(fpm) (acfm) (dscfm)
2,516 19,300 8,120
221 17,000 8,980
2,458 18,600 8,340
219 16,700 9,210
CONTROL PERFORMANCE
Excess
Air
313
193
Evaporation
Rate
(lb/hr)
15,100
11,500
Partlculate Loadlne
(gr/acf) (gr/dscf>
0.0780 0.1854
0.0337 0.0637
0.0673 0.1523
0.0360 0.0654
Pressure Water Avg. Particle Participate Collection
Drop Usage Size
(in. H20) {gptiri (u)
2.8 I25k/
2.9 125^ ~
Rate
(Ib/hr1*
12.9
4.9
10.9
5.2
Efficiency
(7.)
62.0
52.6
§_/ Laboratory result.
b/ Estimated recirculation rate; estimated make-up rate = 2 gpm.
-------
TABLE 26
Koch
(Rozel/Dryer)
KOCH/ROZEL PERFORMANCE AND PROCESS DATA
NO
Test
1
2
4
6S/
Test
1-1
1-0
2-1
2-0
4-1
4-0
Process
Weight
Rate
db/hr)
29,400
29,400
27,400
19,600
Temp.
225
158
242
150
210
147
226
DRYER CONDITIONS
Back -End
Protein
(%>
18.3
_
20.1
18.6
Moisture
(7. by Vol.)
45.7
45.0
47.0
47.3
39.6
41.1
17.1
Chop Moistures
Green
74.6k/
71.9
76.4k/
59.3k/
CARRIER GAS
Avg . Ve 1 .
(fpm)
4,443
4.163
4,587
4,132
4,746
4,166
5,843
(%) Temp.
Dry ( 8F)
6.lk/ 301
4.6k/ 316
8.2k/ 280
8.lk/ 326
CONDITIONS
Flow Rate
(acfm>
32,100
29,500
33,100
29 , 300
34,300
29,500
42,600
(dscftn)
12,400
13,000
12,200
12,600
15,000
14,200
24,800
Excess
Air
(%)
306
90
152
907
Evaporation
Rate
(Ib/hr)
21,500
20,700
20,400
10,900
Par ticu late Loading
(gr/acf)
0.0896
0.0866
0.1075
0.0932
0.0674
0.0495
0.0370
(sr/dscf)
0.256
0.196
0.293
0.218
0.154
0.103
0.059
-------
Koch
(Rozel/Dryer)
TABLE 26 (Concluded)
CONTROL PERFORMANCE
Teat
i-i 1
1-0 -
2-1
2-0
4-1 '
4-0
6-1-'
Pellet Prod.
Rate
(tons/hr)
* 3.94
> 4.31
3.46
4.22
Moisture
Flow
(Ib/hr)
29,100
29,800
30,100
31,500
27,500
27,700
14,300
Pressure
Drop
(in. H20)
6
-------
VO
Air Conditioning Corporation
(Lexington/tH-)
Process
Weight
Rate
(Ib/hr)
TABLE 27
ACC/LEXINGTON PERFORMANCE AND PROCESS DATA
Test
2
3
4
Test
2-1
2-0
3-1
3-0
4-1
4-0
Test
2-1
2-0
3-1
3-0
4-1
4-0
19,200
22,500
22,500
Temp.
246
118
236
103
232
108
Pellet Prod.
Rate
(tons/hr)
2.35
1.95
2.15
Protein
(7.)
24.3
24.5
24.5
Moisture
(I by Vol.)
37.2
13.8
26.2
9.4
30.6
6.7
Moisture
Flow
(Ib/hr)
24,200
9,860
18,800
5,420
22,600
3,670
DRYER CONDITIONS
Back- End
Chop Moistures (7.) Temp .
Green Dry ( °F)
76. 3s-' 5.7f./ 290k/
82.3s./ 7.2£/ 290k/
82. 3£/ 7.2*.' 290k/
CARRIER GAS CONDITIONS
Avg. Vel. Flow Rate
(fpm) (acfm) (dscfm)
4,382 29,300 14,600
3,914 26,500 22,000
4,821 32,200 19,100
3,091 20,900 18,700
4,918 32,900 18,300
2,962 20,100 18,300
CONTROL PERFORMANCE
Excess
Air
m
229
229
229
Evaporation
Rate
(Ib/hr)
14,400
18,200
18,200
Particulate Loading
(gr/acf) (gr/dscf)
0.1434 0.2877
0.0431 0.0519
0.0926 0.1570
0.0372 0.0418
0.1331 0.2388
0.0185 0.0203
Pressure Water Avg. Particle Particulate Collection
Drop Usage Size
(in. H?0) (gpm) (u)
. 405k/ 4.8
405k/ < 1
, 405k/
3 405k/
, 405k/
3 405k/
Rate
(Ib/hr)
36.0
9.8
25.6
6.7
37.5
3.2
Efficiency
(7.)
72.8
73.8
91.5
£/ Laboratory result.
b/ Estimated value.
-------
vo
Ul
Thompson
(Topeka/Total)
Test
1
3
5
6
8
9
10
12
Process
Weight
Rate
(Ib/hr)
24,300
32,800
35,500
36,500
33,200
32,800
31,800
33,700
20,800
TABLE 28
THOMPSON/TOPEKA PERFORMANCE AND PROCESS DATA
DRYER CONDITIONS
Protein
(%)
17.0
16.5
-
18.5
18.6
19.3
21.6
21.4
20.0
Chop Moistures
Green
73.0
73.0
76.0
76.6
76.0
77.7
78.5^
76.2^
74.9^
(%)
Dry
4.8-',
8.5^
h/
8.9-'
9.4^
4.3^
h/
8.3-'
5 8-/
h/
5.8-',
b/
7.2-'
Back -End
Temp.
(°F)
c/
280-
275
278
279
294
310
320
325
330
Excess
Air
a)
224
189
165
165
132
216
183
240
271
Evaporation
Rate
17,400*
23,100
22,350*
27,100
25,870*
21,170*
22,640*
25,200
12,840*
These evaporation rates do not reflect the moisture which was added at the Feeder. The amount
of water added at the Feeder and subsequently evaporated in the dryer is shown on page 71
(Table 22).
-------
Thompson
(Topeka/Total)
TABLE 28 (Continued)
CARRIER GAS CONDITIONS
Test
1-1
1-2
3-1
3-2
5-6
5-7
5-8
6-1
6-2
8-1
8-2
9-1
9-2
10-1
10-2
11-1
11-2^
12-2
12-6
12-8
Sampling
Location
Inlet
Stack
Inlet
Stack
Recycle to:
Primary
cyclone
Drum
Furnace
Inlet
Stack
Inlet
Stack
Inlet
Stack
Inlet
Stack
Inlet
Stack
Stack
Recycle to:
Primary
cyclone
Furnace
Temp.
(°F)
213
209
224
206
211
199
188
222
213
223
204
223
215
223
209
226
228
229
236
245
Moisture
(% by Vol.)
43.0
34.1
38.7
42.9
40.6
42.2
41.8
43.1
28.9
45.3
44.7
42.8
41.5
42.2
41.1
37.2
36.0
38.3
38.8
37.3
Avg. Vel.
(fpm)
5,645
3,071
5,565
3,694
7,121
2,139
3,547
5,430
4,128
5,323
4,619
5,181
4,528
5,221
4,591
5,235
5,352
4,420
6,130
2,540
Flow
(acfm)
44,500
32,000^'
43,800
29,100
3,150
2,620
6,470
42,800
32,500
41,900
36,400
40 , 800
35,700
41,100
36,200
41 , 200
42,200
34,800
2,700
5,000
Rate
(dscfm)
19,100
12,300
19,900
12,800
1,430
1,190
3,220
18,100
17,700
17,100
15,600
17,500
16,100
17,800
16,500
19,200
20,200
16,100
1,220
2,290
Particulate
(gr/acf)
0.136
0.114
0.328
0.470
0.681
0.448
0.314
0.278
0.299
0.211
0.157
0.166
0.082
0.101
0.068
0.148
0.137
0.066
0.135
0.052
Loading
(gr/dscf)
0.318
0.225
0.722
1.066
1.499
0.990
0.681
0.657
0.549
0.519
0.336
0.372
0.183
0.235
0.149
0.319
0.284
0.142
0.299
0.113
-------
Thompson
(Topeka/Total)
TABLE 28 (Concluded)
CONTROL PERFORMANCE
Pellet Prod.
Rate
Test (tons/hr)
1-1
1-2
3-1
3-2 .
5-6 '
5-7
5-8
3.45
4.63
> 4.48
6-1 V , ,,
6-2 J 4'43
tl } 4'18
9-1 L
10-1
10-2 ;
11-1
H-2^/
12-2
12-6
12-8
3.65
f
4.25
' 2.76
Moisture
Flow
(lb/hr)
40,200
23,500
35,100
27,000
2,730
2,420
6,470
38,400
20,100
39,500
35,300
36,500
31,800
36,300
32,200
31,800
31,800
27,900
2,160
3,800
Water
Usage
Feeder
-1
0
4.8
0
12.0
2.4
5.4
0
5.8
(gpm)
Venturi
0
o
0
0
1.9S/
2S/
.-
0
2«/
Average Particle
Size
(n)
-
-
7.1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Particulate
Rate
(Ib/hr)
50.2
33.5
123.0
117.0
18.4
10.1
18.8
102.0
83.2
75.7
48.9
55.6
25.1
35.8
21.1
52.4
49.3
19.5
3.1
2.2
a_/ Without recycle
b_/ Laboratory result
£/ Estimated value.
d/ Volume of water not measured.
e/ Volume estimated
-------
vO
oo
Hell
(Grand Island/Dryer)
Test
i/
5S/
7
9
Test
1-0
2-0-L/
3-0
3-R
4-oa/
5-OS/
7-0
7-R
9-0
9-R
Process
Weigh t
Rate
(lb/hr)
38,100
35 , 800
35,800
39,600
42,200
35,500
40,600
Sampling
Location
Outlet
Outlet
Outlet
Recycle to:
Furnace
Outlet
Outlet
Outlet
Recycle to:
Furnace
Outlet
Recycle to:
Furnace
Protein
(%)
19.1
19.0&/
18.8
-
Temp.
nn_
225
225
226
238
231
214
233
254
243
239
TABLE 29
HEIL/GRAND ISLAND PERFORMANCE AND PROCESS DATA
DRYER CONDITIONS
Back-End Fuel
Chop Moistures (%)
Green
74.0
74.7
74.7
74.2
74.0
73.1
74.6
Moisture
(% by Vol.)
50.9
39.9
53.5
54.5
37.5
42.9
49.9
50.1
51.5
Dry
13.2
11.5
11.5
8.8
10.6
8.3
6.8
CARRIER GAS
Avg. Vel.
(fpm)
1,609
2,838
2,081
3,400
2,416
2,523
1,887
3,335
1,867
Temp. Rate
(°F) (scfh)
290 43,900
278 52,100
295
34,800£/
51,000
44,500 ,
ht
48,000^'
CONDITIONS
Flow Rate
(acfm) (dscfm)
31,000 11,000
54,800 23,800
40,200 13,500
21,500 6,930
46,600 22,400
48,700 21,100
36,400 13,200
21,000 7,350
36,100 12,300
51.8
3,330
21,000
7,160
£/ Without recycle.
b/ Estimated value.
£/ First half = 17,900 scfh; second half = 49,000 scfh.
268
168
103
84
42
Evaporation
Rate
(Ib/hr)
26,700
25,600
25,600
28,400
29,900
25,100
29,500
Particulate Loading
(gr/acf) (gr/dscf)
0.105
0.122
0.138
0.543
0.283
0.304
0.101
0.125
0.116
0.218
0.297
0.281
0.412
1.683
0.589
0.703
0.281
0.359
0.339
0.639
-------
VO
VO
Hell
(Grand Island/Dryer)
Test
1-0
2-oa/
3-0
3-R
4-OS/
5-0^
7-0
7-R
9-0
9-R
Pellet Prod.
Rate
(tons/hr)
5.35
4.83
4.83
5.49
5.93
5.22
5.65
TABLE 29 (Concluded)
CONTROL PERFORMANCE
Moisture
Flow
(lb/hr)
32,000
44,200
43,400
23,200
37,600
44,300
36,600
20,600
36,500
21,500
Avg. Particle
Size
(u)
Particulate
Rate
(lb/hr)
.7
.9
3.0
< 1
7.7
28.0
57.4
47.
99.
113.2
127.1
31.6
22.6
35.7
39.2
-------
TABLE 30
ABILENE PERFORMANCE AND PROCESS DATA
Test
101
102
103
114
115
116
119
141
142
143
144
145
o
o
Test
101
102
103
114
115
116
119
141
142
143
144
145
Process
Weight
Rate
(Ib/hr)
26,400
20,900
18,700
24,200
25,600
26,500
21,500
34,400
28,900
33,200
26,400
22,700
Temp.
186.4
179.3
193.7
195.6
206.
190.
179.
161.6
161.0
139.
169.
.7
.6
,4
.7
.2
171.4
DRYER CONDITIONS AND PERFORMANCE DATA
Pellets
Production Rate
(tons/hr)
4.09
4.08
3.78
4.03
4.43
5.07
3.56
4.60
3.82
5.10
3.35
2.40
Moisture
(% by Vol.)
36.7
27.0
27.8
32.0
33.8
31.4
29.3
30.2
33.4
30.6
31.1
29.9
Protein
(%)
13.4
11.4
12.1
17.2
16.8
17.3
16.9
14.5
16.82
13.84
19.36
19.11
Avg. Vel.
(fpm)
5750
5713
5759
6131
5903
5849
6165
6472
6429
6276
6480
6367
Particulate
Chop Moistures (%)
Green
71.6
64.1
63.1
68.7
67.1
63.5
69.5
75.5
75.8
71.5
77.0
80.3
CARRIER GAS
Flow
(acfin)
39,524
39,274
39,587
42,143
40,579
40,203
42,381
45,747
45,443
44,364
45,810
45,012
Dry
12.8
9.0
5.6
11.8
7.9
8.1
11.8
13.5
10.6
12.9
10.3
8.2
CONDITIONS
Rate
(dscfm)
19,791
22,941
22,378
22,205
20,536
21,527
23,715
26,094
24,792
26,126
25,449
25,345
Rate
(Ibs/hr)
66.8
72.5
73.5
51.0
48.2
31.7
46.0
28.5
30.6
34.9
31.2
32.9
Excess Evaporation
Air
(%)
268.7
317.4
317.4
346.1
399.8
297.8
367.7
309.3
291.5
362.3
297.2
312.5
Rate
(Ib/hr)
17,800
12,600
11,400
15,600
16,400
16,000
14,000
24,700
21,100
22,300
19,600
17,800
Particulate Loading
(er/acf)
.197
.215
.217
.141
.138
.092
.127
.073
.079
.092
.079
.085
(gr/dscf)
.394
.369
.384
.268
.274
.172
.226
.127
.144
.156
.143
.151
-------
TABLE 31
KOCH/OXFORD PERFORMANCE AND PROCESS DATA (1973)
Test
104
105
Test
104
105
Process
Weight
Rate
(Ib/hr)
20,800
23,300
Temp.
152.0
149.7
DRYER CONDITIONS AND
PERFORMANCE DATA
Pellets
Production Rate
(tons/hr)
2.83
3.17
Moisture
(% by Vol.)
28.1
32.5
Protein Chop Moistures (%)
(%) Green
18.4 75.0
18.5 75.0
CARRIER GAS COl
Avg. Vel. Flow
(fpm) (acfm)
453 34,687
445 34,111
Dry
9.8
9.2
IDITIONS
Rate
(dscfm)
20,926
19,390
P articulate
Rate
(Ibs/hr)
Excess Evaporation
Air Rate
(%) (Ib/hr)
9.2 319.8
11.4 282.4
Particulate Loading
(gr/acf) (gr/dscf)
.031
.039
.051
.069
15,100
16,900
-------
o
to
Test
109
110
111
112
113
130
131
132
146
147
148
Process
Weight
Rate
(Ib/hr)
23400
23400
21100
24800
17800
19500
20500
20800
18100
20000
18200
Test
Temp.
109
110
111
112
113
130
131
132
146
147
148
238.6
238.3
248.5
154.0
150.7
242.1
149.1
144.6
151.0
156.1
268.4
TABLE 32
KOCH/LAWRENCE PERFORMANCE AND PROCESS DATA
DRYER CONDITIONS AND PERFORMANCE
Pellets
Production Rate
(tons/hr)
2.75
2.50
2.55
2.50
2.80
2.45
2.45
2.60
2.65
2.42
2.53
Moisture
(% by Vol.)
26.5
36.3
26.5
34.9
26.1
26.6
33.2
31.3
32.1
34.8
28.8
Protein Chop Moistures (%)
(%) Green
17.0 78.6
17.0 80.1
17.0 77.5
18.6 81.5
18.6 70.4
16.3 76.6
16.3 77.5
17.6 76.8
17.77 72.0
18.8 76.8
19.4 73.4
CARRIER GAS
Avg. Vel.
(fpm)
5513
5654
5635
4689
4770
5596
4491
4447
4289
4484
5625
Dry
14.6
17.4
16.7
14.9
12.6
16.9
14.0
15.0
10.1
9.7
11.9
CONDITIONS
Flow
(acfm)
38,969
39,972
39,834
33,150
33,718
39,551
31,740
31,433
30,317
31,694
39,760
DATA
Particulate
Rate
(Ibs/hr)
22.0
21.7
18.8
8.30
6.59
28.5
13.3
12.2
13.74
13.60
41.21
Rate
(dscfm)
20,974
18,661
21,123
18,065
20,932
21,126
17,807
18,243
17,367
17,217
19,926
Excess
Air
(%)
446.7
446.7
446.7
408.3
408.3
503.1
329.1
356.8
421.7
421.7
362.3
Evaporation
Rate
(Ib/hr)
17500
17800
15400
19400
11800
14000
15100
15200
12500
14800
12700
Particulate Loading
(gr/acf)
.066
.063
.055
.029
.023
.084
.049
.045
.053
.050
.121
(gr/dscf)
.122
.136
.104
.054
.037
.157
.087
.078
.092
.092
.241
-------
TABLE 33
THOMPSON/ST. MARYS PERFORMANCE AND PROCESS DATA (1973)
Process
Weight
DRYER CONDITIONS AND PERFORMANCE DATA
Pellets Particulate
Rate Production Rate Protein Chop Moistures (%) Rate
Test (Ib/hr) (tons/hr) <%) Green Dry (Ibs/hr)
120 30,000
121 34,100
122 33,600
Temp.
Test (°F)
120 103.2
121 108.3
122 109.3
3.90 21.5 76.0 4.2 14.7
3.90 20. 7*/ 78.8 6.0 17.1
4.34 20. 7±/ 76.0 5.5 24.5
CARRIER GAS CONDITIONS
Moisture Avg. Vel. Flow Rate
(Z by Vol.) (fpm) (acfm) (dscfm)
44.2 3582 28,211 12,114
46.4 4170 32,843 13,774
44.7 3903 30,740 13,331
Excess
Air
203.9
112.3
205.3
Evaporation
Rate
(Ib/hr)
22,400
26,400
25,100
Particulate Loading
(gr/acf) (gr/dscf)
.061 .142
.061
.093
.145
.214
a/ Protein content of only one sample taken, during these two tests, by the plant operating personnel. The
pellets which were loaded "on stream" into a rail car during the day of these two tests were reported to
contain 17.2% protein.
-------
TABLE 34
NEODESHA PERFORMANCE AND PROCESS DATA
Test
123
124
125
126
127
128
129
Test
Process
Weight
Rate
(lb/to)
26600
24300
23000
19800
21900
19300
15300
Temp.
123
124
125
126
127
128
129
229.1
224.0
237.0
236.9
234.2
238.3
231.0
DRYER CONDITIONS AND PERFORMANCE DATA
Pellets
Production Rate Protein Chop Moistures (%)
(tons/hr)
2.97
2.97
3.20
3.10
2.77
2.97
2.70
Moisture
(% by Vol.)
37.1
32.1
33.7
31.9
35.2
28.5
19.2
(%) Green
17.3 79.1
18.5 76.8
17.7 73.8
17.6 70.6
14.8 76.4
15.8 71.4
16.8 67.0
CARRIER GAS
Avg. Vel.
(fpm)
1911
2116
1951
1923
1931
1938
2113
Dry
14.9
10.0
15.5
13.6
14.1
13.5
15.5
CONDITIONS
Flou
(acfm)
37,204
41,193
37,990
37,437
37,604
37,734
41,153
Particulate
Rate
(Ibs/hr)
28.2
44.0
28.9
35.3
43.4
42.2
23.7
Rate
(dscfm)
17,541
21,059
18,661
18,857
18,158
19,955
24,926
Excess Evaporation
Air
(%)
298.6
386.0
312.5
344.2
312.5
450.7
491.0
Rate
(Ib/hr)
20100
18000
15900
13100
15900
12900
9300
Particulate Loading
(gr/acf)
.088
.125
.089
.110
.135
.130
.067
(gr/dscf)
.187
.244
.181
.218
.279
.247
.111
-------
o
in
TABLE 35
BERTHOUD PERFORMANCE AND PROCESS DATA
DRYER CONDITIONS AND PERFORMANCE DATA
Test
133
134
135
Test
133
134
135
Process
Weight
Pellets
Rate Production Rate Protein Chop Moistures (%)
(Ib/hr) (tons/hr) (%) Green Dry
22,300
21,300
26,500
Temp.
(°F)
213.5
225.5
224.0
3.15 18.2 73.7 10.3
2.05 19.9 82.0 8.4
2.80 19.7 80.5 10.7
CARRIER GAS CONDITIONS
Moisture Avg. Vel. Flow
(% by Vol.) (fpm) (acfm)
36.2 2222 33,408
37.7 2267 34,074
44.4 2312 34,753
Particulate
Rate
(Ibs/hr)
46.5
31.7
45.8
Rate
(dscfm)
14,241
13,910
12,691
Excess Evaporation
Air Rate
(%) (Ib/hr)
387.8
239.7
239.7
Particulate
(gr/acf)
.162
.108
.154
15,800
17,100
20,700
Loading
(gr/dscf)
.381
.266
.422
-------
TABLE 36
HEIL/DUNDEE PERFORMANCE AND PROCESS DATA (1973)
DRYER CONDITIONS AND PERFORMANCE DATA
Process
Weight
Pellets
Particulate
Rate Production Rate Protein Chop Moistures (%) Rate
Test
136
137
138
139
140
Test
136
137
138
139
140
(Ib/hr)
27900
41600
42700
26800
30900
Temp.
(°F)
205.4
221.4
209.3
215.7
214.7
(tons/hr) (%) Green
5.62 14.3 63.0
6.50 18.3 72.2
6.53 15.7 72.2
4.45 13.85 70.0
4.75 14.5 71.5
CARRIER GAS
Moisture Avg. Vel.
(% by Vol.) (fpm)
44.7 1164
46.2 1433
54.3 1413
48.8 1314
50.6 1363
Dry (Ibs/hr)
7.0 12.8
11.0 6.65
11.0 10.8
11.5 6.17
9.6 9.28
CONDITIONS
Flow Rate
(acfm) (dscfm)
22,859 9,417
28,130 11,020
27,752 9,374
25,797 9,651
26,771 9,633
Excess
Air
(%)
171.2
110.2
68.2
159.4
84.8
Particulate
(gr/acf)
.066
.028
.046
.028
.040
Evaporation
Rate
(Ib/hr)
16800
28600
29300
17700
21200
Loading
(gr/dscf)
.159
.070
.135
.075
.112
-------
TABLE 37
DRYER SPECIFICATIONS
Plant
Rossville, Kansas
Arlington, Nebraska
Darr, Nebraska
Neodesha, Kansas
Topeka, Kansas
Grand Island, Nebraska
Oxford, Kansas
Lexington, Nebraska
Rozel, Kansas
Abilene, Kansas
Oxford, Kansas
Lawrence, Kansas
Dundee, Kansas
St. Marys, Kansas
Berth oud, Colorado
Code
AS/
B£/
C£/
na/
E
F
G
H
I
J
K
L
M
N
0
Make of
Dryer
Jones
Hell
MEC
MEC
Thompson
Hell
McGeehee
Thompson
Hell
MEC
McGeehee
Hell
Hell
Thompson
MEC
Size
1040
105
1242
1040
12 x 36
12 1/2 x 42
1034
12 x 36
105
1040
1034
824
125-42
12 x 42
1040
No. of
Passes
1
1
3
3
1
3
1
1
3
3
1
3
3
1
3
Evaporative
Capacity (Ib/hr)
18,000
18,000
30,000
20 ,000
30,000
34,000
12,000
30,000
18,000
22,000
12,000
12,000
34,000
27,000
20,000
af Plants tested In 1971 study (Ref. 1).
-------
TABLE 38
BAY AREA
PROCESS WEIGHT TABLE
MAXIMUM ALLOWABLE EMISSION RATE
Process Weight Rate Rate of Emission Process Weight Rate Rate of Emission
Ib/hr
100
200
400
600
800
1,000
1,500
2,000
2,500
3,000
3,500
4,000
5,000
6,000
7,000
8,000
9,000
10,000
12,000
tons/hr
0.05
0.10
0.20
0.30
0.40
0.50
0.75
1.00
1.25
1.50
1.75
2.00
2.50
3.00
3.50
4.00
4.50
5.00
6.00
Ib/hr
0.551
0.877
1.40
1.83
2.22
2.58
3.38
4.10
4.76
5.38
5.96
6.52
7.58
8.56
9.49
10.4
11.2
12.0
13.6
Ib/hr
16,000
18,000
20,000
30,000
40,000
50,000
60,000
70,000
80,000
90,000
100,000
120,000
140,000
160,000
200,000
1,000,000
2,000,000
6,000,000
tons/hr
8.00
9.00
10.
15.
20.
25.
30.
35.
40.
45.
50.
60.
70.
80.
100.
500.
1,000.
3,000.
Ib/hr
16.5
17.9
19.2
25.2
30.5
35.4
40.0
41.3
42.5
43.6
44.6
46.3
47.8
49.0
51.2
69.0
77.6
92.7
Interpolation of the data in Table 25 for other process weights
shall be accomplished by use of the'following equations:
Process weights < 30 Ton/hr - E=(4.1) (pO-67)
Process weights 30 Ton/hr - E = (55) (P0-11) - 40
Where: E = rate of emissions in Ib/hr
P = process weight in Ton/hr
108
-------
SECTION XI
ACKNOWLEDGEMENTS
The assistance of the Production Committee of the American Dehydrators
Association in developing this project and evaluating the results is
acknowledged, with sincere thanks. The Production Committee consisted
of Mr. Ray Bert, Mr. Robert M. Bunten, Jr., Mr. James T. Healey,
Mr. John Higgins, Mr. Warren Johnson, Mr. Llyle King, Mr. Henry Moeller,
Mr. George Polony, Mr. C.A. Vinci, Mr. H. L. Wells, Mr. Alva Caple,
Mr. Marvin Domina, Mr. Jerry Isaak, and Mr. Ray Lang.
The construction and operation of the Hell recycle system was performed
by Mr. Don Kaminski, Mr. Gordon Llndl and Mr. Lowell Frank of The Hell
Company; and Mr. Paul Newsome of Morrison & Quirk, Inc. The Construc-
tion and operation of the Thompson recycle system was performed by
Messrs. Theodore and Stanley Thompson of the Thompson Dehydrating Co.
The installation and operation of the pilot and full size wet scrubbers
was performed by Mr. Ray Bert, Bert & Wetta Sales, Inc.; Mr. Clifford
Bossung, Dawson County Feed Products, Inc.; Mr. Hank Wells, Oxford
Dehy Company; and Mr. C. A. Vinci, Western Alfalfa Corporation.
Analytical work and preparation of much of the material for this
report was performed by a team from Midwest Research Institute under
the direction of Dr. Chatten Cowherd, Jr., project leader. Mr.
Nicholas Stich and Mr. Pat Shea shared the responsibilities of
directing the MRI Field Crews in the performance testing of control
equipment at the various plant sites and overseeing the monitoring of
process operating conditions. Miss Christine Guenther performed the
mathematical computations of test results. Mr. Paul Gorman conducted
presurveys of the testing sites, and contributed to the evaluation of
control equipment costs and performance data. Mr. Mike Hammons
conducted the laboratory analyses of particulate samples.
Mr. Leslie Hardison, Air Resources, Inc., assisted in evaluating test
data and provided some of the discussion in this report.
Mr. A. Lyle Livingston, USDA, ARS, WRRL, collected product samples and
he and Dr. George Kohler provided the evaluation of the effect of
"recycle" on the quality of dehydrated alfalfa.
Mr. Ray Buergin, Supervisor of Engineering and Enforcement, Kansas
State Department of Health, made the original suggestion for demonstra-
tion of a recycle system which led to the development of this project.
The support of this project by the Environmental Protection Agency
under Research Grant R801446, the assistance of Dr. Bill Garner in
preparation of the grant application and the help provided by Dr. Robert
Bethea, and Mr. Gil Haselberger, EPA Project Officers, in carrying out
the project and preparing this report is acknowledged with sincere thanks.
109
-------
SECTION XII
REFERENCES
1. Cowherd, Chatten. Jr., "Particulate Emissions and Process
Conditions at Representative Alfalfa Dehydrating Mills,"
Final Report MRI Project No. 3538-C (for American Dehydra-
tors Association), November 19, 1971.
2. Federal Register, Volume 36, Number 247, pp. 24876-24895,
December 23, 1971
3. Cowherd, Chatten, Jr., and A. E. Vandegrift, "A Review of
Source Testing Procedures," 21st Sanitary Engineering
Conference, University of Kansas, January 1971.
4. Shigehara, R. T., W. F. Todd, and W. S. Smith, "Significance
of Errors in Stack Sampling Measurements," Annual
Meeting of the Air Pollution Control Association, St. Louis,
Missouri, June 1970.
5. "Source Sampling," National Air Pollution Control Administration
Training Manual. Research Triangle Park, North Carolina
(available to Trainees attending source-sampling course).
6. Shannon, L. J., and P. G. Gorman, "Particulate Pollutant System
Study: Volume II - Fine Particle Emissions," Final Report
MRI Project No. 3326-C (for U. S. Environmental Protection
Agency), August 1, 1971.
7. "Compilation of Air Pollutant Emission Factors," U. S.
Environmental Protection Agency, Office of Air Programs
Publication No. AP-42, February 1972.
Ill
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORT NO
EPA-650/2-74-007
3 RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Particulate Emissions from Alfalfa Dehydrating
Plants -- Control Costs and Effectiveness
5 REPORT DATE
January 1974
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S)
Kenneth D. Smith
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
American Dehydrators Association
5800 Foxridge Drive
Mission, Kansas 66202
10. PROGRAM ELEMENT NO.
1AB012 (ROAP 21ADJ-83)
11. CONTRACT/GRANT NO.
Grant R 801446
1? SPONSORING AGENCY NAME AND ADDRESS
SPA, Office of Research and Development
SERC-RTP, Control Systems Laboratory
Research Triangle Park, N. C. 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final--7/72-1/74
14 SPONSORING AGENCY CODE
15 SUPPLEMENTARY NOTES
16. ABSTRACT ^.^ repOrj presents the results of an extensive field-testing program to
characterize particulate emissions from alfalfa dehydrating plants and to evaluate the
cost effectiveness of available control methods. Results of the first test phase (1971),
performed under a variety of controlled conditions to characterize plant emissions
:rom the various process operations, indicate that the drying operation is responsible
for more than 75% of total emissions. Results of the second test phase (1972), with
benchmark performance data obtained on two pilot-scale and three full-scale wet
scrubbers and on two full-scale recycle systems, indicate that medium-efficiency
wet scrubbers have the potential to bring alfalfa dryer emissions into compliance
with process weight/rate standards. Results of the third phase (1973)--with full-
scale control devices/systems tested to determine: the effectiveness and freedom
:rom operating problems of an intermediate pressure drop scrubber and several
recycle systems, and the effectiveness of plant modifications and operating proc-
edures in reducing particulate emissions--indicate that dehydrating plants can be
operated in compliance with emission standards by using appropriate plant modifi-
cations and operating methods. The report includes capital and operating expenses
for the control schemes investigated, applied to a representative control model.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Croup
Air Pollution
Dehydration
Field Tests
ost Effectiveness
apitalized Costs
Operating Costs
Scrubbers
Air Pollution Control
Alfalfa
Stationary Sources
Particulate Emissions
Emission Standards
13B
07A
14A
8 DISTRIBUTION STATEMENT
Unlimited
19 SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
122
20. SECURITY CLASS (Thispage)
Unclassified
22 PRICE
EPA Form 2220-1 (9-73)
112
------- |