improving semi-dry scrubber performance through gas …improving semi-dry scrubber performance...

ABSTRACT

13th North American Waste to Energy Conference May 23-25, 2005, Orlando, Florida USA

NAWTEC13-3156

IMPROVING SEMI-DRY SCRUBBER PERFORMANCE THROUGH GAS FLOW MODELING

Nikita (Nick) Gorsky Covanta Projects, Inc.

A Covanta Energy Company 40 Lane Road, CN 2615

Fairfield, New Jersey, 07007 Tel. 973-882-7298 Fax 973-882-4168

E-mail: [email protected]

C. F. Peter Bowen, P.E., President Nels Consulting Services, Inc.

40 Neilson Avenue St. Catherines, Ontario

Canada L2M 7M9 Phone: 905-682-2969

E-mail: [email protected]

Poor flue gas flow distribution in the semi-dry scrubbers used in Waste-to-Energy facilities can cause reduced residence time for lime slurry spray droplet evaporation and subsequent "wet carryover" resulting in solids deposits on the scrubber vessel walls and ductwork and also baghouse bag blinding. In addition to promoting corrosion, the removal of deposits during boiler outages is very labor intensive.

This paper identifies how gas flow modeling conducted in conjunction with Nels Consulting Services, Inc. on several different types of scrubbers at Covanta Energy's Waste-to-Energy facilities resulted in modifications which increased the actual flue gas residence time, considerably reduced the solids deposits (scale) and associated maintenance costs, and in some cases reduced the pressure drop across the scrubbers and baghouses.

The data presented includes typical model study velocity distribution data (before and after the modifications), vessel sketches, and photographs. Associated work included in-field scrubber outlet duct temperature and velocity distribution testing. The results of the in-field scrubber outlet temperature distribution testing, done both before and after the scrubber modifications, confirmed the improvements numerically by showing reduced flue gas temperature variation in the scrubber outlet duct.

INTRODUCTION AND SEMI-DRY SCRUBBING THEORY Covanta Energy operates twenty-five (25) waste-toenergy facilities, all of which employ semi-dry scrubbers for acid gas (HCI and S02) control to maintain environmental compliance with all applicable federal, state, and local emission regulations. With the exception of one facility, all facilities are also equipped with fabric filters (baghouses) for particulate control.

In a typical semi-dry scrubber at Covanta, flue gas from exits from the boiler in the range of approximately

81

400 to 475 degrees F and enters the scrubber vessel through a scrubber inlet section. Figure I shows the typical flue gas flow path and Figure 2 shows a typical scrubber/baghouse orientation in a typical Covanta waste-to-energy facility.

Atomized lime slurry containing a suspension of the reagent calcium hydroxide [Ca(OH)2] is sprayed into the vessels. The basic chemical reactions are shown in Figure 3. The actual reactions, however, are much more complex. A gas-liquid mass transfer takes place first whereby the acid gases briefly condense to form aqueous

Copyright © 2005 by ASME

acids which then react with the reagent in a liquid-solid mass transfer process precipitating the products of reaction (solids in suspension). Mathematic model predictions have been compared with experimental results to enable a good understanding of those reactions [ I ].

The lime slurry is diluted with the required amount of water for scrubber outlet temperature control. The resulting lime slurry typically varies between 3% and 20% calcium hydroxide solids by weight, depending on the amount of reagent needed for acid gas control and the amount of water required for scrubber outlet temperature control.

Atomization is accomplished in one of two (2) ways, i.e. with either a rotary atomizer or dual fluid nozzles. Most Covanta scrubbers have only one (I) rotary atomizer per vessel (although there is one facility with a total of three). The remaining scrubbers have multiple

inlet ducts (usually three or four) and have multiple dual fluid nozzles.

Rotary atomizers employ a high speed rotating wheel (in excess of 10,000 rpm) with a means of introducing liquid at its center. The centrifugal force directs the lime slurry to flow horizontally and radially outward through circumferentially located round outlets (nozzles)of the wheel. A film is formed directly at the wheel periphery which quickly disintegrates into very fine droplets [2]. In theory, the droplets briefly travel horizontally in the atomizer wheel area the scrubber vessel, after which they are acted on by the high velocity vertical flue gas flow to form an umbrella-like flow pattern in the vessel. Dual fluid nozzles typically employ compressed air to atomize the lime slurry to very fine droplets and the droplets discharge in a conical pattern, parallel to the vertical flue gas flow.

The scrubber is typically a vertical cylindrical vessel

with a cone bottom. There are basically two(2) types of semi-dry scrubbers, up-flow and down-flow, depending on the direction of flue gas in the vessel. On up-flow vessels, the bottom cone typically serves as a cyclone to remove any fly ash prior to the vessel. On down-flow vessels, the cone either acts as a transition device to a bottom outlet or the cone is used as a transition for any reaction products which by-pass the vessel outlet and settle to the bottom where they are removed by the ash handling system. In this case the gas outlet is horizontal out of the cone.

Figure 4 shows a typical early design down-flow rotary atomizer scrubber introduced in the early 1990' s showing an expected flow pattern of the flue gas and lime slurry and the reaction and drying regions. Figure 5 shows a typical down-flow scrubber vessel with multiple inlet ducts. This design is normally used with multiple rotary atomizers or dual fluid nozzles. Figure 6 shows a typical up-flow scrubber vessel using multiple dual fluid nozzles located above the inlet cyclone. In each scrubber design, there are normally gas dispersion devices (vanes) to guide the flue gas into the vessel. Figures 7 and 8

82

show typical dispersers at the scrubber inlets on the rotary atomizer scrubbers.

Semi-dry scrubbers typically are sized for having enough volume to enable the acid gas chemical reaction to achieve completeness and, moreover, to assure that the lime slurry droplets achieve enough dryness such as not to cause "wet carryover" and subsequent solids build-up in the vessel or the outlet ductwork. To assure that this does not occur, the scrubber suppliers sometimes perform computerized flow dynamics (CFD) analyses to forecast the trajectories and evaporation rates of the slurry droplets as acted upon by the hot flue gas entering the vessel.

In scrubber design, the term "flue gas residence time" is often used, which is typically defined as the volume of the scrubber cylinder divided by the volumetric gas flow assuming plug flow, i.e. assuming that the flue gas has an ideal uniform distribution in the cross-section of the vessel. These theoretical flue gas residence times are typically between 8 to 15 seconds in the industry, although Covant's latest standard scrubber specifications indicate a required minimum of 15 seconds. Earlier up-flow scrubber designs with dual fluid nozzles have design velocities of 5.5 to 6.5 feet per second [3] which translates to approximately 8 seconds residence time.

OPERATING EXPERIENCE Covanta's operating experience over the last 15 years or so has shown that not all of the scrubbers performed as expected, i.e. without any appreciable build-up of lime reaction solids. Heavy build-up of solids was frequently realized between the typical semi-annual facility outage periods on some scrubbers. The build-ups were from the high and non-uniform moisture content of the particulate.

The build-ups were in various degrees in various areas of the scrubber (roof, walls, and outlet ductwork. Once the build-up on the scrubber walls and roof would grow sufficiently, the material would sometimes dislodge and plug the scrubber hopper. In some extreme cases, the pluggage would cause a complete gas flow blockage resulting in the removal of the boiler from service due to loss of boiler draft. Downstream of the scrubber, the build-up would end up on the floor of the scrubber outlet (baghouse inlet) duct, eventually propagating into the baghouse modules.

Besides causing high flue gas pressure drop across the scrubber, ductwork and filter bags (which results in associated energy penalties and lack of good draft control in the furnaces), the build-up has a long term effect of corrosion of the carbon steel platework due to the acidity of the solids and the cold metal temperatures, requiring re-plating or repair within a 5 to 10 year period. The high pressure drop across the filter bags required

higher pulsing pressures and pulsing frequencies which in tum impacted bag life.

Initially, these problems were accepted as inherent in the operation of scrubbers where liquid is sprayed. In fact, in early scrubber designs, some manufacturers


installed devices such as wall scrapers (which cycled on a regular basis to remove wall accumulations) or lump crushers (located at the outlet of the scrubber hoppers to

help the accumulated ash to be removed from the system).

It was initially perceived that the only thing which could be done was to improve atomization, i.e. to achieve a finer droplet size array, thus a higher droplet surface area which would optimize gas-liquid contact and increase evaporation rates. On rotary atomizers, the factors which affect the atomization quality (higher rotational speed, lower liquid flow rate, decreased liquid viscosity [I]) could not be easily achieved, thus no additional optimization could be achieved. On dual fluid nozzles, however, some improvements were realized

after the Sauter Mean Diameter (SMD) of the droplet size array was reduced by installing more efficient nozzles, raising the operating air pressure, and increasing

the air consumption. However, these remedies resulted in only marginal improvements.

It was then decided to investigate possible improvements from a gas flow distribution perspective which lead to the scrubber gas flow model studies. It was perceived that the following was occurring within the vessel:

• the contact and mixing between the flue gas and the spray droplets was poor.

• The flue gas residence time in the vessel was insufficient to dry the droplets of lime slurry.

Nels Consulting, Inc. was contacted to perform the gas flow model testing, based on their experience in flow modeling of various power plant components including scrubbers. In each case, the objective was to optimize the gas flow distribution in the scrubber such that the atomized lime slurry droplets have a longer contact with the hot flue gas. The program involved four (4) phases in each case as follows:

• Initial site visit and field testing. During this phase, the operating parameters and operating experience was reviewed in detail with facility personnel. The testing involved measurements of the gas velocity and temperature distribution in the scrubber inlet and outlet ductwork. This information was evaluated to determine the primary cause of the reported operating problems, location of the problems, and possible optimization strategies.

• The model study itself as described below. • Implementing the recommended modifications

to the scrubber and ductwork design based upon

the model study. • A final site visit and field test of the gas flow

parameters to confirm the success of the modifications.

83

MODEL DESIGN AND CONSTRUCTION In gas flow modelling, it is theoretically necessary to simultaneously maintain geometric, kinematic and dynamic similarity. However, such a condition is not possible under modelling conditions. Therefore, the air flow model studies were conducted using a geometric linear scale factor of 116 or 118 and a velocity scale of Ill. These particular scaling factors were chosen based on the following rationale:

a. When a sufficiently large geometric scaling factor is used the most critical aspect of geometric similarity is satisfied and the reliability of measurement can be ensured. A linear scale of 1/6 or 1/8 was determined to be sufficient to satisfy this criterion and yet economic to build and test.

b. Kinematic similarity is dependent upon the Reynolds Number (Re) and is assured if this number is maintained above an approximate value of 40,000 (i.e., well above the Re of 4,000 which is cited as a typical borderline for turbulent flow conditions). Based on the velocities used in the system, the typical Reynolds Numbers for the full scale installation and the model were in the range of 6 x 105 and 2 x 105 respectively, thus satisfying the above criterion. As fully turbulent flow conditions were maintained in the studies, velocity distributions and pressure losses could be accurately measured in the model and predictions made for the full scale installation.

The models were constructed complete with scrubber inlet and scrubber outlet ductwork. The scrubber inlet and outlet ductwork was constructed from 114-inch Plexiglas and the scrubber vessel from 118-inch Plexiglas. Turning vanes, fabricated from sheet metal, were installed in the ductwork. Specified internal structural members considered significant to the gas flow were included in the models.

MODELING EQUIPMENT AND INSTRU MENTATION The models were operated under positive pressure by a single-inlet centrifugal fan driven by a 30-hp motor. Air flow velocities and pressures in the model were measured with the following equipment:

a. Velocities in the ductwork: Air Flow Developments Mark 5 inclined manometer, range 0 to 0.5, 1, 2, 10 and 20 inches of water with a Pitot static tube (AMCA Std. 210-74, ASHRAE Std. 5175) or'S' type Pitot static tube 15 inches long by 2.3 mm diameter.

b. Velocities in the SDA vessel: a TSI Inc. digital hotwire anemometer, model 8355.


c. Total Pressures: total pressure hole of the Pitot

static tube and manometer as in (a) above.

d. Static Pressures: W' brass wall taps, Pitot static tube static pressure holes and the manometer as in (a) above.

e. Volume Flow: The volume flow through the model section was monitored in the inlet and outlet ductwork where there were sufficiently uniform velocity distributions as indicated by a Root Mean Square Deviation (RMS) of less than 15% of the mean velocity.

f. Air Flow circulation patterns: Powermist Turbo Smoke Generator, #40/80 granulated cork dust, Video camera/recorder, Panasonic AG-455, Ylinch tape, VHS format, and video editor, Panasonic AG-1950

TEST LOCATIONS AND PROCEDURES The typical test locations on a down-flow rotary atomizer scrubber used in the model studies are shown on Figure 9 as T l through T6. A similar scrubber with a side outlet is shown on Figure 10. The velocity and total pressure distributions were measured at each test location within the model. These test positions were selected such that the flow patterns and pressure drops within the ductwork and scrubber vessel could be accurately evaluated.

For evaluation of the pressure losses and velocity distributions through the system, air velocities and total pressure measurements were taken at discrete points located at the centers of equal areas in the duct and vessel test section. The total and velocity pressures were measured using the pitot static tube and manometer,

whereas the air velocities is the scrubber vessel were measured using the hotwire anemometer. A static wall tap and manometer, together with data from the inlet and outlet ductwork, were used to measure pressure loss from the inlet to the vessel and then from the vessel to the outlet ductwork to better estimate field pressure drops due to the change in temperature in the full scale installation.

The data at each test section was presented as a matrix of numbers shown in the location and orientation in which they were recorded, all viewed in direction of flow. The RMS deviation from the mean was calculated and used as a measure of the uniformity at these locations. An example of this data is shown in Figure 11.

Flow direction in the scrubber vessel, as viewed perpendicular to the side wall, was determined by a thin thread attached to the end of the hotwire anemometer. A flow angle reference number, similar to the hour positions on a clock dial, was ascribed to the flow at each measuring point to numerically characterize the flow patterns in the vessel.

For viewing purposes, the smoke generator was used to show the flow patterns at which the air is introduced into the vessel through the roof gas ducts. Also, granulated cork dust, was injected into the inlet ductwork to show gas flow

84

patterns entering the top of the SDA and swirling near the side wall. Videos and photos were made on each modelling project.

MODEL TESTING RESULTS AND MODIFICATIONS Table I lists all of the facilities Covanta operates and identifies which scrubbers were modeled with Nels. A total of nine (9) model studies were done comprising most of the different types of scrubbers. On all of the down-flow scrubbers modeled, the one common result was that there was a high velocity jet of flue gas entering the scrubber at the center of the vessel (on rotary atomizer scrubbers) and at each inlet (dual fluid nozzle scrubbers). On most scrubbers modeled there were also many flow vectors either horizontal (indicating a spin) or even reverse flow (opposite to the predominant gas flow in the vessel). Figure 11 shows a typical pattern on a down-flow rotary atomizer scrubber.

Further away from the vessel inlet and throughout most of the cylindrical portion of the vessel, the jets did not dissipate or slow down much. This was reflected by both velocity array measurements and also the smoke and cork dust injection tests. Furthermore, some vessels

showed extreme turbulence in certain areas. Results from the most recent scrubber modeled (an up-flow design with dual fluid nozzles shown in Figure 6) show that most of the flue gas flow is near the vessel walls in the shape of a hollow annulus. In this case, the model study showed a rather broad reverse flow zone located in the center of the vessel.

The optimization was largely focused on reducing the high velocity jet or jets. As to be expected, the gas flow optimization was done differently for the different scrubbers, frequently involving several iterations of optimization.

On the down-flow single rotary atomizer scrubbers with existing spin vanes in the center, the gas flow was by-passed from the center using equally spaced by-pass ducts and with perforated plate disks as shown in Figures 12 through 15. On similar scrubbers with a tangential inlet and circular, continuously reducing cross-section, inlet scroll and dual cones (inlet design is shown in Figure 8), the by-pass tubes extended from the floor of the inlet scroll through the roof of the vessel as shown in Figure 16.

On the down-flow dual fluid nozzle scrubbers, spin vanes were added at the discharge of each duct as shown in Figure 17. On the up-flow dual fluid nozzle scrubber recently tested, a horizontal perforated plate annulus ring was added in the throat of the scrubber to move most of the gas flow away from the walls as shown in Figure 18. This rather simple modification eliminated the reverse flow in the center of that vessel. In all cases, the modifications eliminated any reverse-flow velocity vectors.

Table 2 provides more insight indicating some of the results (both initial and final), indicating how much improvement was gained in the final optimized


configuration. The improvement was visually confirmed by both the smoke test and the cork dust test.

Modifications were implemented in the field within 4 or 5 days on each scrubber, within the time frame of a semi-annual outage.

OPERATIONAL IMPROVEMENTS, FIELD TESTING, AND CONCLUSIONS Reports from all affected facilities indicated reductions in deposits on the scrubber vessel walls and the scrubber outlet ductwork. Some facilities even reported that all deposits were essentially eliminated. Most facilities experienced a reduction of over-all scrubberlbaghouse system pressure drop and elimination of any evidence of "wet carryover" on the bags.

On rotary atomizer scrubbers, modifications had two (2) additional benefits as follows:

• reduction of the pressure drop across the scrubber by approximately 50% (due to the bypassing of some of the flue gas away from the relatively high pressure drop center disperser).

• reduction of the amount of erosive wear experienced on the existing center disperser vanes.

Improvements in most cases were shown by field tests with reduced flue gas temperature variation across the

85

cross-section of the scrubber outlet duct. Table 3 compares the data for two types of scrubbers.

In conclusion, the improvements showed that there is often not enough residence time for drying the lime

slurry droplets in the scrubber vessels due to the high velocity jets of flue gas and associated turbulence which exists in the scrubbers. The authors strongly agree that if the actual gas velocity in the scrubbers is reduced and better gas distribution is achieved, a better and longer slurry/flue gas contact results and the actual flue gas residence time would not be so far away from the ideal residence time used to size the vessels.

REFERENCES [I] Kolluri, R., Hatcher, William 1. Jr., Modeling Spray Dryer Performance in Flue Gas Scrubbing, 1991, Chemical Engineering Department, The University of Alabama.

[2] Lefebvre, Arthur H., 1989, Atomization and Sprays, published by Taylor & Francis, pages 6, 7, 127 through 135,189, and 223.

[3] Sankey, Mark R. and Licata, Anthony, Air Pollution Control for Waste-to-Energy Plants - What Do We Do Now?, April 17, 1997, Hamon Research-Cottrell Technical Library, Somerville, NJ 08876, page 7


Facility Location No. of Gas Flow Scrubber Atomizer Units Per Type Type

Unit (dscfm@ 7% O2)*

Alexandrial Arlington VA 3 32,726 Downtlow Rotary Babylon NY 2 37,760 Uptlow Nozzles Bristol CT 2 32,726 Downtlow Nozzles Detroit MI 2 109,958 Downtlow Rotary Fairfax County VA 4 75,52 1 Downtlow Nozzles Haverhill MA 2 83,073 Downtlow Rotary Hennepin County MN 2 60,417 Uptlow Nozzles Hillsborough County FL 3 40,278 Downtlow Rotary Honolulu HI 2 85,792 Downtlow Rotary Huntington NY 3 25,174 Downtlow Rotary Huntsville AL 2 34,740 Downtlow Rotary Indianapolis IN 3 73,104 Downtlow Rotary Kent County MI 2 3 1,467 Downtlow Nozzles Lake County FL 2 26,583 Downtlow Rotary Lancaster County PA 3 40,278 Downtlow Rotary Lee County FL 2 60,417 Downtlow Rotary Marion County OR 2 27,691 Uptlow Nozzles Hartford CT 3 67,465 Downtlow Rotary Montgomery County MD 3 60,417 Uptlow Nozzles Onondaga County NY 3 37,257 Downtlow Rotary Pasco County FL 3 35,243 Downtlow Rotary Stanislaus County CA 2 40,278 Downtlow Nozzles Union County NJ 3 48,333 Uptlow Nozzles WaUingford CT 3 14,097 Downtlow Nozzles Warren County NJ 2 22,656 Downtlow Rotary

Table 1. Covanta Energy Facilities and Scrubber Types

* dscfm@7%02 = dry standard cubic feet per minute corrected to 7% Oxygen ** planned

86

Model Modifications Study Type Done

** Yes Spin Vanes

Yes Spin Vanes Yes

Yes By-pass ducts

Yes Spin vanes Yes By-pass ducts

Yes By-pass ducts

Yes By-pass ducts Spin vanes

Yes Annulus ring


Facility Scrubber Atomizer Inlet Ideal Initial Initial Final Final Type Type Gas Mean Actual % Actual 0/0

Flow Vel. Mean R M S Mean R M S (klb/ (ft/ Vel. Dev. Vel. Dev. hr) min) (ft.l from (ft.l from

min.) mean min.) mean

Lake Down Rotary 169 122 1273 527 633 222

Fairfax Down Nozzles 555 293 1063 189 418 107

Union Up Nozzles 347 294 373* 112 306 53

* This mean velocity is low due to the large area of reverse flow in the center

Table 2. Model Study Analytical Results (Different Scrubber Designs)

Down-flow Rotary Downflow Dual Fluid Atomizer Scrubber Nozzle Scrubber

Pre- Post-Mods Pre- Post-Mods Mods Mods

Mean Temperature, of 275 291 308 317 RMS Deviation from Mean, % 6.7 2.3 2.8 1.2 Maximum Temperature, of 289 303 320 322 Minimum Temperature, of 216 280 288 310

Table 3. Comparison of Typical Scrubber Outlet Temperature Distribution Testing

Copyright © 2005 by ASME 87

Boller

Flue Gas Flow Path

Scrubbing is achieved by Injecting lime slurry into the flue gas stream.

10 Fen

Figure 1. Flue Gas Flow Path in Typical Covanta Energy Waste-to-Energy Facilities

Gas Scrubber

Particulale Collectors

D

Figure 2. Typical Semi-Dry Scrubber and Baghouse Arrangement


BASIC CHEMISTRY OF ABSORPTION

H+ -CL

Cft(OH)2+S02 = CaS03+1-t20 +2HCL = ClIICL2+2H20

+HgCL2 ". ClIIHgCL2*H20

Figure 3. Scrubber Chemical Reactions

rotary

nuegas atomizer

from .... boiler

/J

] //1\ \'\. flue gas!

II I \ \\ slurry / I I \ \ \ contact

/ I I 1 \ \ area I I I \ \ \

I I I \ \ \

] I I \ \

I I I \ dryimg I I I \ area

... to baghouse

Figure 4. Typical Early Design Down-flow Rotary Atomizer Scrubber


Figure 5. Typical Down-flow Scrubber with Dual Fluid Nozzles and Multiple Inlet Ducts

Figure 6. Typical Up-Flow Scrubber with Dual Fluid Nozzles


Figure 7. Typical Scrubber Center Disperser Vanes (as modeled)

tone assembly

inlet seroll

innerconc v·.nes

outer cone vanes

7 outer cone

disperser tone iifti lugs

f. rotary atomizer

Figure 8. Typical Tangential Inlet Scroll and Cone Assembly


Figure 9. Typical Model Study Test Locations - Bottom Outlet Scrubber

• x •

Figure 10. Typical Model Study Test Locations - Side Outlet Scrubber


4113

5t2

1094

IIORMAUZI!D VELOCITY. YAW AHOU!

1 2 3 4 5 e 7 a a 10

A 41t 411 485 eae 1012 loe1 111 lID 2M? 1357

B 3211 m 4� 105 2lOS aos 122 1021 2113 13M

C 401 403 SN 141 tQ7" 7.tO 740 14341 2U5 1151 "IW eae an ltT4 2m 7It 711 20SI 1094 582

aOA FLOW ANO� REFlMNCI NUMBeR

1 2 J 4 S 5 7 • • 10

A 2 2 1 1 12 , 1 7 3

I t 1 1 a • T4 T3 • •

C 2 a a a T8 T3 3 I •

C, 2 • • T3 ! J TI 1

3

� 5 1

SM FLOW ANGLE REFERENCE NIJIIItJER

II a

THEOREllCAL MEAN I/!LOCITY- 122 ,PIII lIlIAN VILOCITY- un FPM lUll DEVIATION- UU " 01' MfAN YEl.OCITY STATIC PRUlUR£- 1.51 "MoO GAS TEMPERATUq- U .,. SOA FLOW�"t.V1EW1!O PERPt!NDtCUl.AR To \/usEL WALL

A1 ."

411

46S

II1II

UI12

lID

1357

IZ2

7It

11

122

i21

732 4113

11

1

2 2

12

711

l13li S7I 53S

12

2

43:/

Ii05

1174

m

GAS flOW

Figure 1 1. Typical Pre-Modifications Model Study Results at Location T4


93

�-------- 3 ' --------�

Figure 12. Typical Modifications with Scrubber Inlet By-Pass Ducts

Figure 13. Typical Modifications with Scrubber Inlet By-Pass Ducts (photo)


Figure 14. Model Scrubber Roof Showing the Perforated Plate Disks with Support Plates

Figure 15. Full Scale View of Perforated Plate Disk Looking Into By-Pass Tube


4ClJI:OP("" J>ERf. PLAl£ Ar.Q CONE .o.sS[MBLV

27'-10 '/2�ID

SDA

Figure 16. By-pass Tubes on a Tangential Inlet Down-flow Scrubber

2·9" I .D .�

12 EQUALLY SPACED� 30· VANES (CW SPI N)

Figure 17. Spin Vanes Installed on an Inlet Duct of a Down-flow Dual Fluid Nozzle Scrubber


improving semi-dry scrubber performance through gas …improving semi-dry scrubber performance...

Documents