Flux Force/Condensation Scrubbing For Collecting Fine Particulate From Iron Melting Cupola Draft


AIR POLLUTION TECHNOLOGY, INC.

flux Forcc/Conrfensjticn Scrubber System

For Collection of Fins P.jr ticultle

Emissions rrsn A?> Iron Malting lupoid

hy

Ric^^rd L'. Chm-p!ew.sk* and Ssysou** Ct.»vert

Air Fn"*|-t'C»n TrCunc'c-jj-, Inc.
^901 Mr.»-ara L'lve., Si-•;».«? 402
San Diejo, Cil:fordin 3211/

Contract No. 6Q-02-2H'"i

EPA Project Officer: Pa!"2 I. Par:-tor.

4901 MORENA BLVD., SUITE 402 SAN DIEGO, CA 92117 (714) 272 0060

* .	No-tics ,

¦ "T!iis Cjouasnt is a preli~i nri**;*1 t. DRAFT

' It lias not been formally rcl::~osI t/ LJ'A and

c'.-.culcl r.ot at this stage bo ccr.str... *2 . .<	i-A

«*» represent A^sricy policy, I 'i 11, isir -t ci. --	APT

lated ~or oocaaent on it a toc...dcal Ljcur.. /	_ 5>

arid policy implications.	G*»

-------
LIST OF FIGURES, Continued

Figure	Page

7-28	Predicted and Experimental Penetration Run

37/3	94

7-29	Predicted and Experimental Penetration Run

37/4	95

7-30	Predicted and Experimental Penetration Run

37/5	96

7-31	Predicted and Experimental Penetration Run

37/6	97

7-32	Predicted and Experimental Penetration Run

37/7	98

7-33	Predicted and Experimental Penetration Run

37/8	99

7-34	Predicted and Experimental Penetration Run

37/9	100

7-35	Predicted and Experimental Penetration Run

37/10	101

7-36	Predicted and Experimental Penetration Run

37/11	102

7-37	Predicted and Experimental Penetration Run

37/12	103

7-38	Predicted and Experimental Penetration Run

37/13	.	104

7-39	Predicted and Experimental Penetration "Run

37/15	105

7-40	Predicted and Experimental Penetration Run

37/16	106

7-41	Predicted and Experimental Penetration Run

37/17	107

7-42	Predicted and Experimental Penetration Run

37/18	108

7-43	Predicted and Experimental Penetration Run

37/19	109

7-44	Predicted and Experimental Penetration Run

37/21	110

7-45	Predicted and Experimental Penetration Run

37/23	111

7-46	Predicted and Experimental Penetration Run

37/24	112

7-47	Predicted and Experimental Penetration Run

37/25	113

7-48	Predicted and Experimental Penetration Run

37/26	114

-------
LIST OF FIGURES, Continued

Figure
7-7

7-8

7-9

7-10

7-11

7-12

7-13

7-14

7-15

7-16

7-17

7-18

7-19

7-20

7-21

7-22

7-23

7-24

7-2S

7-26

7-27

Predicted
36/4

Predicted
36/5

Predicted
36/6

Predicted
36/7

Predicted
36/8

Predicted
36/9

Predicted
36/10

Predicted
36/11

Predicted
36/12

Predicted
36/13

Predicted
36/14

Predicted
36/15

Predicted
36/16

Predicted
36/17

Predicted
36/18

Predicted
36/19

Predicted
36/20

Predicted
36/21

Predicted
36/22

Predicted
37/1

Predicted
37/2

and Experimenta
and Experimenta

and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta

and Experimenta

*

and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta
and Experimenta

and Experimenta

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration-Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Penetration Run

Page

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

-------
LIST OF FIGURES

Figure	Page

3-1 Particle Size Distribution Site One	9

3-2 Particle Size Distribution Site Two	10

3-3	Particle Size Distribution Site Three	11

4-1	F/C Demonstration Plant Flow Diagram Induced
Draft	15

4-2	F/C Demonstration Plant Flow Diagram Forced

Draft	16

4-3 Induced Draft System (Plan View)	17

4-4 Induced Draft System (Elevation View)	18

4-5 Forced Draft System (Plan View)	19

4-6 Forced Draft System (Elevation View)	20

4-7 Process Flow Sheet Induced Draft System	21

4-8 Process Flow Sheet Forced Draft System	.22

4-9 Photograph of Saturator and Condenser	25

4-10 Settler	26

4-11 Condenser, Scrubber and Cooling Tower	26

4-12 Exhaust Fan Wheel	26

4-13	Fan Wheel Detail	31

5-1	Modified EPA Sampling System with In Stack
Cascade Impactor	45

6-1	Simplified F/C Scrubber System	47

6-2 Example of Initial and Grown Size Distri-
bution	55

6-3	Example of Condenser, Scrubber and Combined
Penetration	56

7-1	Histogram of Condenser Inlet Mass Concentra-
tion	67

7-2 Average Condenser Inlet Size Distribution	68

7-3	Average Condenser Inlet Cumulative Size

Distribution	69

7-4	Predicted and Experimental Penetration Run

36/1	70

7-5	Predicted and Experimental Penetration Run

36/2	71

7-6	Predicted and Experimental Penetration Run

36/3	72

-------
LIST OF TABLES

No.	Page

3-1	Summary of Alternate Cupola Operations	7

4-1	Pump Specifications	33

4-2 Flow Measurement Specifications	40

6-1 Input Variables For Prediction Of Scrubber

Performance	53

6-2	Example Calculation	54

7-1	F/C Scrubber System Operating Parameters

For 36 Series	60

7-2 F/C Scrubber System Operating Parameters

For 37 Series	61

7-3 Gas Flow Rates, Temperatures and Measured

Condensation Ratio For 36 Series	62

7-4 Gas Flow Rates, Temperatures and Measured

Condensation Ratio For 37 Series		 63

7-5 Particulate Mass Concentrations	64

7-6	Particulate Mass Concentrations	65

8-1	Summary of Equipment Costs F/C Scrubber

System	118

8-2 Summary of Equipment Losts Conventional

Scrubber	119'

8-3 Direct and Indirect Cost F/C System	121,

8-4 Direct and Indirect Cost Conventional

Scrubber	122,

8-5 Summary of Power Requirements	123,

8-6 Summary of Annual Operating Costs	123;

-------
TABLE OF CONTENTS

I.	INTRODUCTION 			 . . . 1

II.	SUMMARY, CONCLUSIONS, RECOMMENDATIONS. ........ 3

III.	SITE SELECTION			6

IV.	SYSTEM DESIGN. 			 . .14

V.	PERFORMANCE TESTING			40

VI.	SYSTEM PERFORMANCE MODEL 	 ........ .46

VII.	COMPARISON OF EXPERIMENTAL RESULTS AND PERFORMANCE

MODEL					.59

VIII.	ECONOMIC ANALYSIS. ....... 	 ... 115

-------
I

SECTION 1	.

INTRODUCTION

Flux force/condensation (F/C) scrubbing has been developed by
Air Pollution Technology under EPA-sponsorship for the past several
years. The objectives of (F/C) scrubbing involve' reducing the
power requirement for collection of fine particulate compared to
conventional high energy scrubbers. The improvement in scrubber
performance due to condensation effects is most apparent in the
sub-micron size range where the mechanism' of inertial impaction is
difficult to apply economically, f . tj> "*

J

There are three condensation effects which are utilized in
F/C scrubbing. The condensation of water vapor from a hot satura-
ted gas can be caused by contacting with a colder liquid. The
suspended particles in the gas act as condensation nuclei resulting
in growth of the particulate due to condensation of water vapor.
The transfer of water vapor toward the cold liquid results in
diffusiophoresis. This mechanism collects particles by movement
of the condensing water vapor toward the cold surface. Simulta-
neously, the temperature gradient established in the condenser
results in a thermal force which results in collection by thermo-
phoresis; The particle growth due to condensation plus the flux
forces of diffusiophoresis and thermophoresis enhance the particle
collection efficiency of the F/C scrubber system.

This report presents the results of a demonstration project
to verify the design methods and economics of flux force/condensa-
tion, scrubbing on a large scale in an industrial environment. The
F/C system was designed, built and tested on the total exhaust gas
stream from a 12,500 Kg/hr iron melting cupola. The operating
experience involved seven months of operation. This included
summer, fall and winter in a demanding environment resulting in a

-------
good test of the equipment design and materials selection.

Performance tests were made over a period of several months;.
The test results verified the design methods and showed that the
F/C scrubber system could meet all emission limits at a lower
power requirement than a conventional high energy scrubber. An
economic analysis showed that a F/C scrubber system has a lower
annual operating cost compared to a high energy scrubber.

-------
SECTION 2

SUMMARY, CONCLUSIONS, and RECOMMENDATIONS

SUMMARY

The objectives of this program were to demonstrate the per-
formance and economics of a flux force/condensation scrubber
system in an industrial environment. The maior tasks	I

were:

1.	Select a company in the iron and steel industry suitable
for the demonstration scrubber.

2.	Design the demonstration scrubber system for the chosen
site.

3.	Prepare a test plan.

4.	Fabricate, install and start up F/C system.

5.	Conduct test program.

6.	Maintain laboratory capability to support test urogram.

« - -

7.	Engineering—analysis—oi system performance.

8.	Site restoration, if needed.

9.Recommendations	for a future demonstration plant.

All of the required tasks were completed. Since the system
has been appraised and purchased by the host company, no site
restoration was required.

A survey of three iron melting cupolas revealed that the
main process parameters of exhaust gas temperature, flow rate and
particulate concentration varied from site to site. Even for
similar sized cupolas the gas flow rate, temperature, particle mass
concentration, and size distribution varied. The site chosen had
the largest production rate and uncontrolled mass emission rate of
the three investigated.

The process weight limit required that the overall system
penetration be less than 0.16 (161) based on measured emission rates

-------
after the spray saturator. The mass mean aerodynamic diameter was
found to average about 1.2 jiimA. The F/C scrubber system required
a total power requirement of 233 Kw (299 HP) versus 345 Kw (462 HP)
for a conventional high energy scrubber. This power savings was
obtained by (1) reducing the necessary pressure drop for the
F/C system and by (2) reducing the volumetric flow rate of the
exhaust gas which occurs during the F/C process due to cooling.

The total annual operating cost for an F/C system operating
on this source was estimated to $84,400 compared to* $111,500 for a
high energy scrubber.

CONCLUSIONS

The following conclusions can be drawn from the results of
this study:

1.	The performance data verified design methods developed by
A.P.T. on previous EPA contracts.

2.	The F/C scrubber system was built, operated for 6 months,
and shown to be in compliance with all applicable air pollu-
tion regulations for that site.

3.	The F/C scrubber system required only about 651 of the
power requirement of a conventional high energy scrubber
to achieve the same performance.

4.	The emissions from the cupola varied in terms of particle
mass concentration and size distribution during the charg-
ing cycle. In this regard continuous or semi-continuous
charging operation with conveyor system would probably
help reduce the variation in emissions which was observed.

5.	The particle number concentration is very important in
F/C scrubbing. For a given amount of condensation the
number of particles which receive the condensed water
determine the grown particle size.

6.	The solids/liquid separation presented some maintenance
problems and suggests that additional attention be given
to equipment design, both to prevent settling in the
scrubber equipment and to improve settling in the water

-------
treatment system.

7.	Corrosion can be a serious problem in this type of appli-
cation. Sulfuric and sulfurous acids from the sulfur in
the coke as well as carbonic acid must be neutralized in
order to prevent significant deterioration of steel compon-
ents in the system. Maintenance of the proper circulating
water pH is also important if stainless materials are used
because of the chloride content present.

8.	Operation over a 6 month period showed that the F/C system
is capable of performing well and without significant diffi
culty in the demanding conditions of a ferrous foundry.
During the entire test period there were no production
stoppages due to the F/C scrubbing system.

9.	Incidental benefits for both plant operation and air
pollution control came as a consequence of the detailed

. engineering study of the cupola operation.

-------
SECTION 3
SITE SELECTION

ALTERNATIVE SITES

The site selection for the F/C demonstration plant involved
preliminary investigation of three foundry operations. The first
candidate was a foundry department in a shipyard which produced
grey iron castings for use in the shipbuilding activities. The
cupola had an internal diameter of 0.94 m and a melting rate of
approximately 2,700 kg/hr (3 ton/hr). This cupola was operated
with an iron to coke ratio which ranged from 5.6 to 7.2 depending
upon the type of material being charged. The average iron to
coke ratio was about 6.

This foundry operation produced a number of castings which
required close attention to the composition of the metal produced.
As a consequence each charge of coke, flux, iron and steel was
weighed before charging and close control of melt rate and tempera
ture was maintained.

Molten metal was taken from the cupola into a ladle,which
was used to pour the castings. This resulted in intermittent
withdrawl of metal from the cupola, consequently the tuyere blower
was operated in an on and off manner. The result was a wide
variation in exhaust gas flows, temperature and particle concen-
tration.

Table-3-1 summarizes the operating parameters for this foundry
as well as two additional foundries which were evaluated. Parti-
culate sampling tests were performed at a location downstream of
the hot gas quencher.

The second foundry produced cast iron pipe and fittings.
The cupola was significantly larger having an internal diameter

-------
TABLE 3-1. 1 SUMMARY OF ALTERNATE CUPOLA OPERATIONS

Site 1

Site 2

Site 3

Cupola Diameter (m)

Charge Wt. (kg)

Melt Rate (kg/hr)

Iron to Coke Ratio

Tuyere Air (DNm3/s)

Exhaust Gas Flow (DNm3/s)

Exhaust Gas Temperature (°C)

•Exhaust Gas Particle

Concentration (g/DNm3)

0.94
438
2,700
6
1.0
2. 5
600-950
1.0

UL */*>\ Z'01-

•Measured after quencher

1.67
682
10,250
7
2.5
' 5.8
350-1,050
1.5

&.40



1.67
1,815
12,500
6

3.5

6.6
350-1,100

1.8

_£SS
1*3 .

u M

/C j / 5 XcP, j /O

9

3,3



: Sr

V

4

-------
of 1.67 m and a melt rate of approximately 10,250 kg/hr. Table
3-1 shows the process conditions for this cupola. The iron to
coke ratio for this cupola averaged about 7. The charges were
weighed before being loaded into the cupola. The tuyere air
flow rate was controlled with an orifice meter, damper and an
automatic controller.

The molten metal from this cupola was withdrawn continuously
into a holding ladel from which metal was taken for various
casting operations. This continuous operation resulted in proces
conditions which were more uniform.

The third site investigated was a foundry which produced
metal shot. The cupola had an internal diameter of 1.67 m.
The production rate varied from about 8,200 kg/hr to 13,500 kg/hr
depending upon the charging rate. The metal was withdrawn con-
tinuously from the cupola. Neither the charge weight nor the
tuyere air flow rate was measured regularly by the foundry.per-
sonnel.

PARTICULATE DATA FROM PRELIMINARY TESTS

Particulate,sampling tests were performed at each of the
three sites primarily to obtain particle size distribution data.
Sampling was performed using the University of Washington cascade
iiapactors. A few Method 5 total filter runs were made to deter-
mine particle concentration and mass flow rate. Figures 3-1, 3-2,
and 3-3, show the particle size distribution measured with the
U.W* impactors at the three sites. In each case the sampling
was performed at a location downstream of the spray quenchers
which were used to cool the hot cupola exhaust gas. The design
of the quench systems varied from plant to plant, however, each
of the quench systems utilized preformed sprays,which probably
resulted in some collection of larger particulate matter.

The particle size distribution measurements at each site
were found to be reasonably consistent from run to run. At the
first site the average mass mean particle aerodynamic diameter
was found to be about 0.45 pmA. The average mass concentration

-------
<

m
i->
m
is
<
1—1

>-
O
O
OS

W
<

1-4

OS
<
a,

6
5
4

0
0

95 98 99

20 30 40 (50 '60 70 80

WEIGHT I
-------
5 10 20 30 40 50 60 70 80 90 95
WEIiHT PERCENT < dp

Figure 3-2.

Particulate Size Distribution Site Two

-------
E

S3.

o
u

i—«

se
Ss

a
o

Q£
UJ

-------
was 1.0 g/DNm3. This was the smallest cupola tested and the
scrap charge was weighed very closely. The combination of small
cupola and a closely controlled charge may contribute to the
small particle size distribution measured. For the second site
the average mass mean particle aerodynamic diameter was 0.60 pmA.
The average particulate mass concentration was 1.5 g/DNm1. The
third site was found to have a mass mean particle aerodynamic
diameter which averaged 0.90 ymA. This large cupola had the
highest particle concentration of 1.8 g/DNm9. No weighing of
the charges was performed at this foundry. The third site had
the largest charge weight as shown in Table 3-1.

SITE SELECTION

The site finally chosen for the demonstration plant was
site 3. The first site proved to be unacceptable when the produc-
tion schedule dropped to six hours per week. Discussion with the
plant management indicated that while there was considerable
interest in the project the production level was not expected
to increase in the near future. The size of the cupola was
relatively small and atypical of large industrial-sources. These
two factors resulted in a decision to seek an alternative host
company.

The second site was a better candidate for the demonstration
system. The cupola was significantly larger and the production
schedule was five days per week. The management of the plant
allowed preliminary testing in order to obtain design data, how-
ever, A.P.T. could not convince the owners of the foundry to
continue with the demonstration project. The lack of available
space for the proposed equipment was one contributing factor
to the decision to reject the second site.

The third site was found and proved to be the better choice
for building the demonstration plant. This site had a cupola
which is representative of the larger cupolas in operation. The
production schedule is normally three ten-hour days per week.

-------
SECTION 4
SYSTEM DESIGN

INTRODUCTION

This section describes the system process design,.the
component design functions, and the structural details of the
F/C scrubbing demonstration plant. The system flow diagrams are
shown in Figures 4-1 and Figure 4-2 for the induced draft and
forced draft configurations, respectively. Both configurations
draw the hot gas from the cupola through the breeching and after-
burner tank to the saturator and condenser vessels. In the induced
draft system the blower is located downstream of the scrubber;
while in the forced draft mode the blower is located between the
condenser and the scrubber. Figures 4-3 and 4-4 show the system
layouts for the induced draft system. Figures 4-5 and 4-6 show
the layout for the forced draft system. Note that only the duct-
ing is modified to switch from forced to induced draft; the remain-
ing components remained stationary.

PROCESS DESCRIPTION

Figures 4-7 and 4-8 show the process flow sheet for both the
induced draft and forced draft configuration. The flow rates and
temperatures are shown for the design condition which would maxi-
mize the F/C effects.

The cupola converts scrap iron and steel to the molten state
using heat generated by combustion of coke. The combustion air
is supplied to tuyeres in the bottom of the cupola by a tuyere
air blower. The cupola is a cylindrical vessel, brick lined
internally and water cooled externally. The melting rate is
determined by the charging rate and tuyere air flow rate. Raw
materials are composed of scrap structural steel, engine blocks,
metal stamping and turnings, cast iron pipe and fittings and other
miscellaneous metal parts.

-------
The foundry had an existing air pollution system consisting of an
afterburner tank, quencher, scrubber and exhaust fan. The existing
system was 12 years old and capable of only marginal performance.
As a consequence the plant management had a keen interest in hosting
the demonstration plant.

-------
f

Figure4-1. F/C demonstration plant flow diagram Induced draft.

-------
1

Figure 4-2. F/C Demonstration plant flow diagram forced draft

-------
AFTERBURNER

SPRAY SATURATOR

CONDENSER

STACK

INDUCED DRAFT SYSTEM (plan view)

Figure 4-3

.(Jim m

-------
ONDENSER

STACK

SCRUBBER

INDUCED DRAFT SYSTEM (elevation view)

Fifiire 4-4

-------
AFTERBURNER

SPRAY SATURATOR

CONDENSER

STACK

SCRUBBER

FORCED DRAFT F/C SYSTEM (plan view)

Figure 4-5

-------
ONDENSER

=9k £=

SCRUBBER

FORCED DRAFT F/C SYSTEM (elevation view)

Figure 4-6

-------
?-3 Scrubber Punp

P~t Hater Trutaant" Pwp

process rum

6AS

Mill Now Rate
pry) lb/sin

Mux Flaw litt
(Vat) Ib/aia

Co=7osicion

0, Ib/ala (lt» wl/ain)

Ki Ib/xda (lb Ml/sin)

CO Ib/ntin (lb ml/min)

COj lb/aia (lb aol/ain)

HiO lb/Bis (lb BOl/aia)

Tot*l (lb aol/Bin)

T«s? *F

frcsiim, in. Hg

CA Fresiur*, la. W.C.

*Cf>!

SC7X (Mac)

8SC7*

TOO

710

792
782

32S
333

183(5.09)
S 57 (19.2)

18(0.36)
(24.6S)

33
•42
9,315
18.215
9, MO

29.11

-0.5

S(.2t)

(12.77)

29.92
8

5,130
4,940
4.130

1.120 1,120

1.135

76(2.31)

S37(19.2) 232(6.97)

107(3.62}

146(3.36)

1.6(0.56)

(26.94)

1.120
1.60S

1,120
1,245

1,120
1.245

1,120
1,245

1SJ.47) 15(.47) 13(.47) 15(.47) 1S(.47)

719(26,2) 719(26.2) 769(26.2) 719(26.2) 769(28.2)

316(7.16) 316(7.16) 316(7.11) 316(7.1*) 316(7.18)

1S(.64) 483(26.9) 125(6.94) 125(6.94) 125(6.94)

(36.7) (62.6)

1,600
29. IS
-1

60,700
14.200
13,900

170
29.77
-2

30,200
13.900

(42.6) (42.6)

136
29.41

-7

20,000
13.900

130
27.57

-32

21,300
13,900

(41.6)

130
29.98
~1

19,600
13,900

UOUfP

JSL

•13

Mas* FJow late, lb/hr

620

ISO

9,110

2,950

6,960

75-ISO

9,430

1,100

360

1.080

10-20

1,136

Te=per*tra* *F

170

ISO

130

100

ISO

hfllOM, ptli

200

20-30

Figure 4-7.

Process

Flow

Sheet Induced Draft System

-------

Cap
Ufttr

ni«t,

•urntr

Cupola After-

kurn*r

. 1

P-l Saturstor Puap •

f-J Coellni To>»r Pap

f-J Scrubber Fuap

-.P-4 Vacar Trmiml *u»p

-pj®| | ^ —

r-3

Hater
Tr*al*cnt

Mass Flow Res*

(Dry) Ib/eie

Mass Flow Sat*

(Ktt) lb/oin . .

Covpositiofl

Oi lb/sin {lb aol/ain)

Hi lb/nin (lb awl/aift)

CO lb/nin (lb mol/mln)

COa lb/fdn (lb Bol/ain)

RtO lb/sin (lb wl/Bin}

Total (lb eol/nin)

T«* *F

ftHMn, In. Hg
CA Pressure, In. V.C.
ACFM

SCFH (Wet)

BSCFK

PUKESS FU0HS

' 4 ' S

•

700

792

328

1,120 1.120

1,120

710

792

333

1,133

1.60S

1,245

1.24S

163(5.09)

76(2.38)

1S(.47)

IS(.47)

li(.47)

IS (.47)

1S{.47)

537(19.2)

S37(19.2)

252(8.97)

789(28.2)

7(9(2t.2)

7(9(21.2)

7B9(2(.2)

789(28.2)

107(3.(2)

- '

141(3.36)

316(7.18)

318(7.18)

316(7.18)

10 CO. Si]

1.6(0.Si)

S(.2«)

1S(.(4)

48S(26.9)

125(6.94)

(34.15)

P6.94)

(12.77)

<36.7)

{62.•)

(42.1)

(42.8)

(«.«)

1,100

170

130

29.8$

29.92

29.1$

29.77

29.41

31.24

29.99

•42

-o.s

-1

-2

-7

~1«

MIS

«.J»

•0,700

30,200

20,000

11,800

19,600

10.2?$

4,940

14,200

9,340

4,130

13,900

LIQUID

1»

•toss Nov Hat*, lb/mM\

620

ISO

•,110

2,990

»,960

75-150

9,430

1,100

360

1,080

10-20

1,136

TcoperattfTe *F

•0

170

ISO

130

100

150

Pressure, piig

200

20-30

Figure 4-8. Process Flow Sheet Forced Draft System

-------
'The scrap and coke are charged together in 1,800 kg charges
(2 tons) with an iron to coke mass ratio of 6. Small amounts of
limestone, 20-30 kg, are also charged to act as a fluxing agent
in the formation of slag. The nominal melting rate is 12,500 kg/
hr. The tuyere air flow rate can be varied by means of a damper
on the blower. The normal tuyere air flow rate is 3.5 DNm3/s.

The exhaust gases from the bed of material in the cupola are
composed of carbon monoxide, carbon dioxide and nitrogen. The gas
analysis is typically 141 CO, 131 C02, and 731 N2. Additional
combustion air at a rate of 2.3 DNm3/s enters the charging door
to allow for complete conversion of carbon monoxide to carbon
dioxide. An ignition burner is installed to ensure that a flame
source is available for combustion of the CO to C02.

Combustion is completed in a brick-lined afterburner tank.
The exhaust gases leave the afterburner tank at a temperature
ranging from 800 - 1»000°C. The exhaust gas flow rate at this
point is about 6.6 DNm3/s.

The hot gases leave the afterburner and are cooled in the
saturator by means of water sprays. The gas is cooled and the
temperature approaches the adiabatic saturation temperature which
depends upon the hot gas temperature. Typically, the saturation
temperature achieved ranges from 71°C to 77°C. The maximum water
spray rate in the saturator is 4.7 £/s (75 gpm).

The saturated gases next enter the condenser which further
cools the gases causing condensation of water vapor. The cooling
capacity of the condenser depends upon the rate of cooling water
flow and the cold water temperature which depends upon the cooling
tower performance. The condenser and cooling tower system has a
design heat rejection rate of 9.6 * 10* J/s (33 * 10s BTU/hr).
The water flow rate can be varied up to 69 L/S (1,100 gpm). The
cold gas temperature leaving the condenser is about 55°C at design
flow rates. The condenser is a counter current packed bed.

Particle collection occurs in the condenser by the mechanisms
of inertial impaction, diffusion, thermophoresis and diffusiophoresis.

-------
K<*

V.f

The condensation of water vapor on the particles results in sub-
stantial growth of the submicron particles. This growth phenom-
enon is a primary aspect of F/C scrubbing which differentiates
this process from other scrubbing applications.

The exhaust gases leave the condenser and enter either the
exhaust fan or the scrubber, depending on the ducting configuration.
The system is designed to allow testing with the scrubber either
in the induced draft or forced draft mode. The scrubber is the
primary particle collection device in the system. The design is
a gas atomized scrubber with variable area throat. The design
incorporates an integral water sump and entrainment separator.
yt The exhaust fan provides the draft for the system. The fan

\Awas designed and built to A.P.T. specifications and is a straight
^1 radial blade design. The flow rate of the fan is approximately

f' 1114 Ams/s (30,000 cfm). at - 96 cm W.C. (- 38 in. W.C.).

The cooling tower cools the water to be pumped into the con-
y-~ denser. It is designed to cool 69 l/s (1,100 gpm) of water from

66°C to 38°C using ambient air. This is accomplished by evaporative
cooling.

The settler is designed to remove particles collected by
the scrubbing liquid. A 1.3 l/s (20 gpm) stream of sump water
is clarified by the settler, The thickened sludge (10-301 by
wt.) is disposed of as land fill. The clarified water (100 ppm *
solids) is returned to the sump.

MECHANICAL DESIGN

The major F/C scrubber system components which are listed
below will be described in detail;

1. Afterburner 5. Venturi scrubber 9. Piping § ducting

2. Saturator 6. Settler 10. Instrumentation

3. Condenser 7. Fan 11. Structure 5

4. Cooling tower 8. Pumps foundation

Photographs of the system and its components are shown in Figures
4-9 through 4-12.

-------
-------
Figure 4-10 Settler

Figure 4-11 Condenser
Scrubber and Cooling Tower

Pioiirp A -1 7 Prti aiKt Fan Wheel as revised bv A.P.T.

-------
Afterburner

The afterburner vessel is a cylindrical steel shell 7.3 m
high with a diameter of 2.4 m. The top third of the afterburner
which was added in a plant expansion is made of stainless steel
while the lower portion is carbon steel. The shell is lined with
7.6 cm of .castable insulation. Gas from the cupola enters the top
portion and flows out the bottom into the saturator. The ignition
burner is located in the duct coming out of the cupola before the
afterburner vessel. The ignition burner is a Hauck gas pilot size
#3 and operates on 0.057 m3/min (2 cfm) of natural gas at a pressure
of 15.24 cm W.C. (6 in. W.C.). A 2.54 cm pipe tapped off a foundry-
gas line is used. The burner unit also requires a flow of air of
about 0.7 m3/min (25 cfm) which was tapped off the tuyere air
fan with a 3.8 cin pipe.

Saturator

• The saturator is a horizontal cylinder rolled from 3.5 mm
(10 gauge) 316 stainless steel sheet. The saturator has a straight
section and an elbow section which connects to the afterburner
tank. The cylinder is 9.1 m long and has a dimeter of 1.5 m.
The vessel is built in three sections and connected by flanges.
The elbow section of the vessel is welded to the afterburners
while the other end forms an expansion ioint at the condenser
inlet. The section of the saturator leaving the afterburner
is insulated with firebrick because of the high temperatures
at that location.

The spray pattern in the saturator is obtained by the
arrangement bf spray nozzles. The spray banks are designed to pro-
vide easy access to the nozzles. Three nozzles with radial sprays
are located in the first (elbow) section of the saturator. A
spray header with 4 nozzles transverses the duct. Forty nozzles
are located on 12 header pipes in the straight section of the
saturator. Three or four nozzles are mounted on 2.54 cm diameter
stainless steel pipes. These pipes can be removed from outside
for maintenance. A flow switch has also been installed on the
main water supply line to the nozzles. When the flow to the

-------
saturator nozzles decreases below 75.8 £/min (20 gpm), the switch
activates ail alarm system and shuts off the main exhaust blower.
This is to prevent high saturator temperatures from damaging the
spray bank and saturator gaskets.

The mechanical design of the saturator was based on two con-
straints. First, the vessel has to have enough volume so the
residence time is sufficient for evaporation of the water drops.
The second constraint involves the available space for construc-
tion of the F/C system. The primary equipment area was to be loca-
ted north of the existing sump, hence, a long horizontal spray
chamber was designed to connect the afterburner with the condenser.

Condenser

The condenser vessel is a 316 stainless steel cylindrical
shell, which is 2.75 m in diameter and 6.1 m high. The wall
thickness is 3.5mm (io gauge). The gas enters the bottom of
the condenser from the saturator and is distributed by turning
vanes.. Counter current contact is made with cold water in a
packed bed having a packing depth of 1.2 m. Intalox rings of
3.3 cm nominal diameter are used as packing. This packed bed
is irrigated by water falling from a trough type distribution
plate. Some of the water droplets are entrained in the rising
gas and a second packed bed, 0.3 m deep, is used as an entrainment
separator. The gas flows out the top of the condenser through
a 0.8 m diameter duct to the scrubber. The water drains down
to the condenser reservoir at the bottom of the condenser and
is drained through a 0.3 m diameter pipe back into the sump.

Scrubber

The scrubber consists of a cylindrical shell 2.1 m in dia-
meter and 2.7 m high. The scrubber shell was rolled from 3.5 mm
(10 gauge) stainless steel sheet. The shell houses both a gas
atomized scrubber section and the entrainment separators.

The inlet ducting is bolted to the gas atomized scrubber
section which in turn is bolted to the scrubber shell. The scubber

-------
section has a variable area throat with hydraulic or manual con-
trol which allows the area to be varied from 0.08 m2 to 0.22 m2.
Scrubber water is pumped to six distribution headers located
just upstream of the converging section of the scrubber. The
gas is directed downward through the scrubber. Scrubber water
is collected in the bottom of the scrubber shell and recirculated
to the scrubber throat. •

Drops of water generated in the scrubber throat are entrained
in the gas stream. To remove these drops entrainment separators
are required at the outlet to the scrubber shell. A tube bank
entrainment separator is used in this application. The entrain-
ment separators consist of six rows of 4.1 cm stainless steel
tubes spaced 4.1 cm from center to center. These rows are offset
to force the gas stream to its way through the tubes be-

fore leaving the scrubber 1 ng. The entrained drops impact
on the tubes and drain to the scrubber reservoir. The entrain-
ment separator cross-sectional area is 5.1 m2 and is made of
1,080 tubes 76.2 cm long.

Cooling Tower

The cooling tower consists of a 3.7 m x 2.7 m x 5.2 m epoxy
coated steel shell. The tower is a standard Commercial
design. Ambient air is forced through the packed tower

a fan. Hot water is distributed across the top of the tower
packing by means of multiple headers and spray nozzles. Cdld
water temperature is controlled by an automatic adjustment of
the air flow.by means of a fan damper.

The cooling tower is designed to cool 69 l/s of water from
66°C to 33°C. The cooling tower incorporates a baffle type
entrainment separator to remove entrained drops from the exhaust
air.

Solids are removed from the circulating water system by
means of a gravity settler. The settler which was used had
a settling area of 11 m2. The design incorporates parallel

Settler

-------
plates inclined at a 55°C angle to minimize the required
area for the unit.

The cleaned liquid (overflow) flows upward along the plates
and overflows through flow distribution orifices. The solids
collect on the plates and slide downward into a sludge hopper.
The sludge is discharged periodically through a sludge valve and
disposed of as landfill.

A flocculant aid is used to increase the settling rate of
the solids. A chemical feed system involving a mix tank and
metering pump supplies the flocculant at a dose rate of 5-10
ppm. The settler design incorporates a flash mix tank and a
floe growth tank as well as agitators with speed controllers.
The nominal flow rate through the settler is 1.3 1/s. The settler
reduces the solids content of the circulating water from several
thousand ppm to a few hundred ppm.

Blower

During the initial phase of the project the blower from
the existing scrubber system was used to provide flow for the
F/C system. The fan then had a straight radial blade design and
was driven by a directly coupled 400 HP electrical motor at
1,800 rpm. The current demand was excessive; requiring 480-500
amps at 480 volts. The performance of the fan was poor and
several modifications were made to obtain the required flow.

Early testing indicated that significant leakage of ambient
air was occurring through the fan housing resulting in reduced
flow through the F/C scrubber system. The entire inlet section
of the fan housing was rebuilt which solved the air leakage
problem allowing design flow through the F/C system. The fan
still had poor efficiency and the developed head was excessive
for the requirements of the F/C system.

Finally, a new fan wheel was designed in order to obtain
more flow and a lower head. The new fan wheel also had straight"
radial blades, but the wheel diameter was r'educed to 1.47 m from
1.60 m and a new blade profile was designed to conform closer
to the fan housing. Figure 4-13 shows a new wheel design. The

-------
I

tut g*T

«> #*•*»

"*««. >** »• * i •# «t*a v*>

K* WIS*-

It lfcrf^ "««*» " ti*

m nuviioM TtcMMoioo*. *k

mssr

r*.M UJKtll.

w»«-

*¦¦«~•• I 3 T- SOO-7

Figure 4-13. Fan Wheel Figure

-------
new fan wheel was installed and successfully operated. The
desired flow rate was attained at a lower head and the motor
power requirement dropped to 360-380 amps at 480 v.

Pumps

Table 4.1 shows the pump specifications for each of the pumps
used on the demonstration plant. All of the pumps are of a proven
commercial design and presented no unusual operating problems.

The saturator pump supplies city water to the spray nozzles
in the saturator. A two stage centrifugal pump is used to pro-
vide the high pressure required to give fine atomization for
effective evaporation.

The sump pump causes the circulation of water from the
sump to the cooling tower. A vertical turbine pump with the
suction submerged in the sump was used for this purpose.

The scrubber pump circulates the water from the bottom
of the scrubber to the scrubber throat. A single stage centri-
fugal pump is used for this purpose.

Two diaphragm pumps are used in the system. The smaller
pump is used to transport sump water to the settler. The larger
pump is used for maintenance purposes such as cleaning condenser
and scrubber, etc. These pumps are portable, driven from com-
pressed air available at the plant and are capable of pumping
liquids as well as sludges with a high solids content.

-------

TABLE 4.1 PUMP SPECIFICATIONS

Pump

Type

Capacity

Driver

P-l Saturator Pump

Z-stage centrifugal

54 Jt/s « IS

atm

20 Hp, 240 V, 30 motor

P-2 Sump Pump

Vertical turbine

750 £./s « I

atm

30 Hp, 240 V, 30 motor

P-3 Scrubber Pump

Centrifugal

250 Jt/s e 1

atm

5 Hp, 240 V, 30 motor

P-4 Water Treatment
Pump

Diaphram

1.3 l/s 8 1

atm

Compressed Air

P-5 Sludge Pump

Diaphram

2.6 1/s 8 1

atm

Compressed Air

-------
Piping, Ducting, and Valves

The piping system is composed of four separate subsystems:
(1) saturator piping, (2) cooling water piping, (3) scrubber
piping and (4) water treatment system piping.

Saturator piping

City water is used in the saturator to cool the exhaust gas.
The water is piped to the saturator pump (P-l) and then passes
through one of two parallel strainers which remove solids which
might plug the spray nozzles. Galvanized pipe having a diameter
of 6.4 cm (2 1/2 in.) is used up to the individual spray headers.
Flexible copper hoses (2.5 cm diameter) are used to connect to
the saturator.

Cooling Water Piping

The cooling water piping runs from the sump pump (P-2) to
the cooling tower, from the cooling tower to the condenser and
from the condenser to the sump. A bypass line around the cool-
ing tower is also included. The material used for the cooling
water system is schedule 40 fiberglass pipe 20 cm diameter (8 in.).

Scrubber Piping

The scrubber piping runs from the scrubber internal sump to
the gas atomized scrubber throat. Fiberglass pipe having a
diameter of 10 cm (4 in.) is used. At the scrubber throat 6
header pipes each 5 cm (2 in.) in diameter distribute the scrubber
liquor. A bypass line allows the scrubber water to be pumped
to the settler.

Water Treatment Piping

The settler system is connected to the circulating water system
by 5 cm diameter PVC piping.

-------
Valves

The valves used in the system were generally for two purposes
(1) flow control and (2) shut-off. The scrubber liquid flow was
controlled using a rubber lined diaphram pump. Ball valves were
used on the saturator spray lines for on-off control. Butterfly
valves were used in other locations for control or shut-off.

Ducting

The ducting used on the project was Fiberglass reinforced
polyester (FRP) ducting. The ducting sections between the
condenser and scrubber are 0.81 m diameter. The exhaust gas
ducting is 0.91 m diameter. The nominal wall thickness is
0.5 cm. The ducting system was designed to allow operation
in both the forced draft and induced draft mode.

Foundations and Structural Design

A reinforced concrete foundation was required to support
the condenser and scrubber vessels. The saturator and scrubber
pumps are also installed on this foundation. The condenser
foundation is a concrete block 3.5 m square by 1.8 m deep. The
adjacent scrubber and pump foundation is a rectangular pad
3.5 m wide by 4.7 m long and 0.5 m deep.

-------
A structural steel platform was required to support the cool-
ing tower. The platform also included a sampling area just above
the scrubber where particulate sampling could be performed both at
the scrubber inlet ducting and at the scrubber outlet ducting. ,, '

INSTRUMENTATION AND CONTROL SYSTEM

/v!' '

Instrumentation —•-C.

The process instrument diagram is shown in Figure 14-14.
Temperatures were measured by means of chrome1-alumel thermocouples
(K type). The temperatures were recorded on a strip chart recorder
and observable on a digital indicator with a 20 point selector
switch.

The liquid flow rates are measured with venturi meters,
transducers, and indicators as shown in Table 4-2. Liquid pressures
are measured with standard pressure gauges, both locally at the
pumps and remotely in the instrument trailer.

The gas pressures and differential pressures acorss the con-
denser and scrubber were measured by means of Minihelic and Magniheli
gauges. Pressure tubing was used between the equipment and the
instrument trailer where the gauges were located.

Saturator Process Measurements

a. Gas inlet temperature (TR-5)

b. Gas outlet temperature (TR-6)

c. Liquid flow rate (FM-12, FI-12)

d. Liquid temperature (TR-12)

e. Pump suction pressure (PI-12.1)

f. Pump discharge pressure (PI-12.2)

g. Strainer discharge pressure (PI-12.3)

Condenser Process Measurements /

a. Gas inlet temperature (TR-6) ^ ^

b. Gas outlet temperature (TR-7) 1 ; '* •

c. Gas inlet pressure (PI-6) \ ! %•

d. Gas outlet pressure (PI-7)

e. Condenser total pressure drop (PI-7/8)

f. Entrainment separator pressure drop

g. Liquid flow rate (FM-14, FI-14)

^ -

-------
h. Liquid inlet temperature (TR-16)

i. Liquid outlet temperature (TR-18)

Cooling Tower Process Measurements

a. Liquid flow rate (FM-14)

b. Liquid inlet temperature (TI-14)

c. Liquid outlet temperature (TI-16)

Scrubber Process Measurements

a. Gas inlet temperature (TI-8)

b. Gas outlet temperature (TI-9)

c. 'Liquid flow rate (FM-15)

d. Liquid temperature (TI-15)

e. Pump discharge pressure (PI-15)

f. Scrubber total pressure drop (PI-8/9)

g. Entrainment separator pressure drop

h. Scrubber liquid level

Water Treatment System Process Flows
a. Liquid flow rate (FM-25)

Control System

Saturator Water System

The saturator water flow rate is controlled by manually
opening or closing the shutoff valves located on the spray bar
headers. The saturator water system also contains two liquid
"Y" strainers piped in parallel to remove particulate which could
clog the spray nozzles. Pressure gauges located upstream and
downstream of the strainers monitored pressure loss across
the strainer.

-------
Cooling Water Flow ^ ,

The cooling water low rate is controlled by means of the by-
pass valve V-14.1. The cooling tower can be bypassed by opening
the cooling tower bypass valve V-28 and closing the cooling tower
valve ¥-14,2. This allows water to be pumped directly to the con-
denser from the sump.

Scrubber Liquid Flow

The scrubber liquid flow is controlled by means of the flow
control valve V-15. In addition the scrubber bypass valve V-23
allows scrubber liquid to be pumped to the sump or to the sewer.

Scrubber Liquid Level 1

The scrubber liquid level is controlled by means of a level
controller which actuated valve V-21.1 on signal of low level.

Scrubber Flow Area

The scrubber flow area can be controlled by a hydraulic
system which opens and closes the scrubber throat area. In addi-
tion, manual adjustment is possible using extension levers located
outside the scrubber.

Gas Flow Rate

The gas flow rate is controlled by a combinatioiT~bf fan inlet
damper and scrubber throat area.

Water Treatment Flow Rate

The flow rate to the water treatment system is controlled by
the air regulator used on the diaphragm pump.

Circulating Water pH

The pH of the circulating water system is measured using a
pH meter and controlled between the range of 6-9 using additions
of sodia ash.

Safety System

The safety system protects the equipment from excessive tem-
perature in the event that the saturator sprays or pump fails to
provide enough cooling water. The saturator water flow rate is

-------

monitored by flow switch (FI-12). If the flow rate decreases
below 25 gpm the flow switch will shut down the exhaust fan and
initiate automatic opening of the cupola cup.

A backup safety system involves a temperature switch (TC-7)
located in the ducting section coming out of the condenser. In
the event gas temperature at this point exceeds the setpoint of
80°C (175°F) the exhaust fan will be shut down and the cupola cap
raised.

Both of these safety switches provide an alarm and indicator
light to alert operators of the malfunction.

-------
TABLE 4.2 FLOW MEASUREMENT SPECIFICATIONS

Identification

Type

Indicator

FM-12
FM-14

FM-15
FM-25

Venturi meter 6.25 cm
cast iron, flanged

Venturi meter 20 cm
cast iron, flanged

Venturi meter 10 cm
cast iron, flanged

Venturi meter 5 cm
brass, threaded

Ammeter 0-20 mA

Voltmeter 0-5

Magnehelic 0-60
in W.C.

-------
SECTION 5

j,J '

PERFORMANCE TESTING ( J

. -v

. 1 i \

i 1

After the installation of the scrubber equipment, instrumenta-
tion, piping and electrical were completed the F/C system was
started up and performance test data/obtained. Start up presented
no problems with equipment supplied for the project. The original
exhaust fan did cause excessive power consumption and^elec-

A' *

trical overload t-**^. This condition continued until the new fan,
which A.P.T. designed^was installed. The system performance was
quite good during the test period. No lost production time was
experienced by the foundry due to problems associated with the
operation of the F/C scrubber system.

Operating Modes

The F/C system was tested in two basic configurations,
induced draft and forced draft mode. In the more traditional
induced draft system the exhaust gas flows through the after-
burner, saturator, condenser, scrubberfand then the exhaust
fan to the stack. By using ducting modifications the F/C
system was tested in the forced draft mode, wherein the exhaust
gas flows through the afterburner, saturator, condenser*^than
through the exhaust fanf^and finally through the scrubber and
stack. This operating mode has the advantage of reducing the
fan power requirement.

The major operating parameters which were varied during tfte
test program were the condenser liquid flow rate and scrubber
pressure drop. The condenser liquid flow rate affected the amount
of cooling in the condenser and hence the condensation ratio.
The scrubber pressure drop could be varied by changing the scrubber

-------
throat flow area and the scrubber liquid to gas ratio. The
scrubber pressure drop determines the collection efficiency in the
scrubber.

Test Locations

Sampling was performed at three locations in the F/C system.
At each of these locations two sampling ports located 90° apart
were available. The sampling ports were made of 10 cm (4 in) pipe
couplings.

The sampling location was at the cold end of the saturator
just before the condenser vessel. The sampling ports at this
location were easily accessible from the ground. The remaining
two sampling locations were in the scrubber inlet and outlet
ducting. These sampling ports were accessible from a sampling
platform which was constructed as part of the cooling tower plat-
form.

Testing Methods

This section describes the test procedures that were used to
determine the F/C system performance. Sampling was usually per-
formed at the inlet to the condenser and at the scrubber outlet.
For some runs sampling was also performed at the scrubber inlet.
Testing at three locations allowed the condenser performance and
scrubber performance to be evaluated separately,

Impactor Runs

The particle size distribution and mass concentration was
obtained using University of Washington Cascade Impactors. Pre-
cutters were used on the impactors to remove large particles and
entrainment. Fiberglass and greased aluminum substrates were used
in the impactors.

Modified EPA Method 5

Several modified BPA Method 5 total filter runs were made to
determine total.mass concentration. These tests were done for .
compliance tests and to confirm impactor mass loadings.

-------
Sampling Equipment

The sampling equipment used involves four sampling trains as
shown in Figure 5-1. The cascade impactor runs were made in stack.
The precutters used had cut points in the range of 6 to 9 vmA, depend
ing upon sampling flow rate. Nozzle sizes were chosen to maintain
isokinetic sampling. The filter tests were made with heated
filters out-of-stack and heated sampling probes. The probes were
made from 0.95 cm (3/8") stainless steel and incorporated S-type
Pitot tubes. Cyclones were not used in the method 5 testing.

Data Analysis

Cascade Impactors

The cascade impactor tests provide data on the cumulative '
mass concentration at specific particle sizes or cut points. From
this data penetrations as a function of particle size could be
determined. The data analysis utilized was a computer program
developed by Dr. Leslie Sparks. The program calculated
impactor cut points and cumulative mass loading based on weight
gain per stage, impactor flow rate and impactor calibration data.

The overall penetration was calculated from the total mass
concentrations as follows. ¦——

Ft « Ftc x Ptg

(5-1)

(5-2)

(5-3)

where Ft « overall penetration condenser and scrubber
Ft^ » overall penetration condenser
Ptg ¦ overall penetration scrubber

CCI » particle mass concentration condenser inlet, g/Dnm

tl*

Cgj - particle mass concentration scrubber inlet, g/Dnm3

-------
CgQ « particle mass concentration scrubber outlet, g/Dnm^

The penetration as a function of particle size was evaluated
in two ways, each of which is based on the following equation:

d(_

inlet

(5-4)

where

is the slope 6f the cumulative mass concentration

curve at a specific particle aerodynamic diameter.

The graphical method involves plotting the cumulative mass
loading versus particle diameter curves for both the inlet and out-
let and determining the slopes of these curves at various points.
The fractional penetration curve is then plotted on the size range
of interest by performing the divi sion shown in equation (5-4) at
each specific diameter required.

The second method used,known as the discrete or finite
difference method,involves approximating the slope of the cumula-
tive mass concentration curve by use of the individual impactor
stage weight gain and cut points, i.e.

(

AC
Alnd_

inlet

/ AC '
IAlndp

(5-5)

outlet

This method is described by McCain et al. (1978) and has the advantage of
being easily adapted to computer calculations. It is not subject
to errors due to determining the slope of the cumulative mass
concentration curve.

-------
I—0—

* PR!

STAINLESS

STEEL
PROBE

PRKCUTTER

ORIFICE METER

CASCADE
IMPACTOR

DRY GAS
METER

VACUUM
PUMP

SILICA

GEL

DRYER

Figure 5-1. Modified E.P.A. sampling train with in stack cascade impactor.

-------
SECTION 6
SYSTEM PERFORMANCE MODEL

INTRODUCTION

The system performance model predicts the particle removal
efficiency in the flux force/condensation scrubber system. The
model can be used to predict the emission from a F/C scrubber
system installed on a pollutant source with a known particle con-
centration and size distribution. The model allows one to vary
independently the two primary variables which influence the per-
formance of the F/C system. These two variables are the condensa-
tion ratio and scrubber pressure drop.

Under previous EPA contracts, Air Pollution Technology, Inc.
conducted detailed studies on the technical and economic feasibili-
ties of F/C scrubbing (Calvert et al., 1973, '75, *77). These
studies included theoretical development of design equations for
F/C scrubbing as well as experimental verification of on both pilot
plants and demonstration plant equipment. The model^used in this
project also uses scrubber performance methods developed in the
scrubber system study (Calvert, 1972) and recent revisions for gas
atomized scrubbers (Yung, 1978).

BASIC CONCEPTS

Before proceeding to the details of the mathematical model,
the basic concepts and outline of the approach will be discussed.
If we consider a typical F/C scrubbing system, it might have the
features shown in Figure 6-1. The gas leaving the source is hot
and has a water vapor content which depends on the source process.
The first step is to saturate the gas by quenching it with water.
This will cause no condensation if the particles are insoluble, .
but will if they are soluble. There will be a diffusiophoretic
force directed away from the liquid surface.

-------
CLEAN GAS

WATER —>

SATURATED
GAS

WATER —D>

SATURATED
GAS

HOT GAS

Figure 6-1.1 Simplified F/C scrubber system.

-------
Condensation is required in order to have diffusiophoretic
deposition, any growth on insoluble particles, and extensive growth
on soluble particles. Contacting with cold water or a cold
surface is employed to cause condensation. While condensation
occurs there will be diffusiophoretic and thermophoretic deposition
as well as some inertial impaction (and, perhaps, Brownian diffu-
sion). The particles in the gas leaving the condenser will have
grown in mass due to the layer of water they carry.

Subsequent scrubbing of the gas will result in more particle
collection by inertial impaction. This will be more efficient
than impaction before particle growth because of the greater
inertia of the particles. There may be additional condensation,
depending on water and gas temperatures, and its effects can be
accounted for as discussed above.

The mathematical model for the F/C demonstration plant is
outlined below:

Saturator

a. Gas is humified and cooled to adiabatic saturation tempera-
ture.

b. No condensation occurs (particles are assumed insoluble).

c. Particle collection is assumed to be negligible in the
saturator.

Condenser

a. Particles are collected by impaction in packed column.

b. Condensation occurs causing growth of particles.

c. Collection occurs in condenser due to diffusiophoresis.

Scrubber

a. Grown particles are collected by impaction in scrubber.

b. Negligible condensation occurs.

PREDICTION OF PENETRATION

Collection by Impaction in Condenser .

The particulate collection in the condenser is primarily due
to inertial impaction. The inlet (dry) particle size distribution

-------
is used for these calculations (no credit is taken for growth for
collection within the condenser). The penetration is calculated
with the following equation for collection in a packed bed.

Ptd ¦ exp

it 2

2(j+j ) (e-Hd) d(

Ur d.

(6-1)

where K ¦ impaction parameter « ——

P 9"g dc

u^ « superfacial gas velocity, cm/s
uG - gas viscosity, g/cm-s
dj> * packing diameter, cm

Z « packing depth, cm

j * channel width fraction, dimensionless

e » void fraction, dimensionless

¦ liquid holdup fraction, dimensionless

d_0 = particle aerodynamic diameter, cm (g/cm3)1/2
pa

For the F/C demonstration plant the condenser was packed with
5.0 cm intalox packing to a depth of 122 cm. The void fraction
was 0.97 and the channel width fraction was estimated to be 0.15.

Condensation and Growth

The growth of the particulate.,due to condensation of water
vapor is predicted based on the humidity entering and leaving the
condenser. The grown particle size is calculated from equation
(6-2).

6 f,

d *
P 2

p q1

+ d

772 it np p

(6-2)

-------
where dp^ « initial particle (physical) diameter, cm

d » grown particle (physical) diameter, cm
P 2

f • « particle condensation fraction

ql ¦ condensation ratio, g l^O/g Dry Gas

np ¦ particle number concentration, particles/cm3

The grown particle density is then calculated by the
following equation:

PP2* f^") ("P"'1) *1-0 C6_3)

where p - initial (dry) particle density, g/cm3

The grown particle density allows one to calculate the final
grown particle aerodynamic diameter with the following equation:

dpa - dp < Cl Pp

where

dp » particle physical diameter, cm

C1 = Cunningham slip correction factor, dimensionless
Pp = particle density, g/cm3

Collection by Diffusiophoresis

Collection by diffusiophoresis in the condenser can be calcu-
lated by means of the following equation:

PtCD - 1 - 0.85 (fy.) Cl-fp)

(6-5)

-------
where * mole fraction of gas condensing

The total penetration in the condenser is then calculated from
the product of equation (6-1) and (6-5).

PtC * PtCI x PtCD

(6-6)

Collection by Impaction in Scrubber

The collection of the grown particulate in the scrubber is by
inertial impaction. The scrubber penetration is calculated according
to the method developed by Yung et al (1978) for collection in a
venturi throat. The penetration equation utsd is as follows:

In Pt,
1

K U* « 0.7
po

4Kpo U*1,3 + 4,2 u*°'5 " 5.02

x fU* +

Po)

0.7

tan

U* K _\0,5
if po

IT-

K + 0.7

(6-7)

x 1 4 K_ + 4.2 - 5.0 2 K°i I 1 + ] tan"1

where B ¦

»l

).7\

Vo)

QG °G CDo

9uC dd

U* - 1.0 - 211 - x + (X* - x

3t CD PG
and x - ..x j " + 1
16 dd PL

)»]

^L/^G - liquid to gas volumetric ratio, dimensionless

CDo * drop drag coefficient at throat entrance, dimensionless

-------
* drop diameter, cm

Ud • drop velocity at throat entrance, cm/s
t - throat length, cm

The penetration in the F/C system can then be obtained from
the product of the condenser and scrubber penetrations;

Pt - Ptc x Pts (6-8)

Pt » PtCI x PtCD x Ptg (6-9)

A computer program has been written to perform these calcula-
tions. The program allows prediction of the scrubber performance
as a function of particle size. The input required for the pro-
gram is listed in Table 6-1 along with numbers for a sample calcu-
lation.

Figure 6-2 shows the initial and grown size distributions for
the sample case. Figure 6-3 shows the penetration as a function
of particle size for the condenser, gas atomized scrubber and the

combined system. The penetrations are plotted vs dry particle
size. The actual scrubber penetration is based on-the grown
particle size distribution as shown in Table 6-2.

-------
TABLE 6-1. INPUT VARIABLES FOR PREDICTION
OF SCRUBBER PERFORMANCE

gas velocity through condenser
void fraction of packing
liquid hold up fraction
packing depth
packing diameter
particle condensation fraction
particle number concentration
particle density
condensation ratio
scrubber throat length
scrubber liquid to gas ratio
scrubber throat velocity

180 cm/s

0.97

0.0

122 cm
5 cm
0.15

1 x 10* particles/cm5
3.0 g/cm3
0.15 g H20/g D.S.
13 cm
0.002
5,000 cm/s

-------
TABLE 6-2. EXAMPLE CALCULATIONS

Initial Particle Grown Particle Penetration, fraction
Aerodynamic Diameter Aerodynamic Diameter Condenser Scrubber Total
pmA ymA Ptc Pts Pt

0.40

0.93

0.88

0.18

0.16

0.50

0.93

0.81

0.18

0.16

0.60

0.95

0.86

0.18

0.15

0.80

1.02

0.85

0.14

0.12

1.00

1.10

0.83

0.12

0.10

1.25

0.81

0.07

0.06

1.50

0.78

0.04

0.03

2.00

0.72

0.02

0.01

3.00

0.56

0.01

0.005

5.00

0.25

0.01

-0

10.00

0.01

-0

-------
10.Q

az
w
H

<
Q

3E
<
35
>•
a
o
ps
w
<

10 20 30 40 50 60 70
Mass I < dpa

80 90 95

Figure 6-2. Example of initial and grown size

distributions.

-------
4-*

63
U
U
©

W
2

W
(X.

9&
98

40
30

20
10

0.3 0.4 0.5 1.0 2.0 3.0 4.0 5.0 10.0

dp, iimA

Figure 6-3 Example of Condenser, Scrubber, And Combined Penetration

-------
Calculation of Particle Number Concentrations

The particle number concentrations were calculated from the
size distribution and mass concentration assuming a log normal
fit to the lowest three impactor cut points. The particle size
distribution at the condenser inlet was bimoda 1 probably due to
a very small condensation aerosol and somewhat larger particles
of ash, sand and soot carried with the exhaust gas to the scrubber
system. If the smallest three cut points are used to construct
a log normal size distribution the proper mass median particle
diameter is generally obtained.

The particle number concentration is related to the mass
concentration and mass mean diameter by the following equation:

"p " * Pp V (6-10)

where

C » particle mass concentration, g/cm
dm ¦ mass mean diameter, cm

The mass mean diameter is related to the mass -median diameter
by the following equation:

In dm - In dpg - 1.5 In og2 (6-11)

where

dpg * mass mediam diameter, cm

0 ¦ geometric standard deviation
g

For several runs the number concentration calculated in this

manner was found to be excessively high. For example aerosol

9 3

systems with number concentration in excess of 10 particles/cm
can not exist ever for short periods of time before agglomeration
of individual particles begins to reduce the particle number

-------
concentration. Coagulation theory describes the rate of particle
agglomation as:

dt p

where

K ¦ coagulation constant

dD- - - K„.2 <«"»>

For iron oxide which is the major constituent of the cupola

particulate, K can be estimated to be 1 x 10 cm /s at the

after burner temperature of 1200°K (Green and Lane, 1964).

Integration q£ equation (6-12) for various initial particle

concentrations shows that a steady state particle concentration is

reached after a few seconds independent of the initial particle

concentration. The residence time of the exhaust gas to the

saturator is about two seconds.

An upper limit particle number concentration can than be

estimated at 5 x 10 particles/cms. This limit was applied to

the previously calculated number concentrations when the initial

8 3

number concentration was found to exceed 5 x 10 particle/cm .

-------
SECTION 7

COMPARISON OF EXPERIMENTAL RESULTS AND
PERFORMANCE MODEL

TEST OPERATING CONDITIONS

The primary F/C scrubber system operating parameters of con-
denser liquid flow, scrubber pressure drop and condensation ratio
are shown in Tables 7-1 and 7-2 for the two months of optimim per-
formance. The "36 series" of tests was made in the induced
draft configuration while the "37 series" was run in the forced
draft mode. 1

The gas temperatures, flow rate and the measured condensation
ratio are shown in Tables 7-3 and 7-4 for these runs. The hot gas
temperature leaving the afterburner was for most of the runs signi-
ficantly lower than anticipated. The design-basis hot gas tempera-
ture was 1,000°C. The maximum temperature measured was about 850°C.
The reasons for the low gas temperature was CI) excessive air leak-
age through the cupola charging door and (2) the pilot burner was
not available for installation until after testing was completed.

The lower than expected hot gas temperature resulted in lower
gas humidity and a consequent reduction in the condensation ratio
which could be achieved. The maximum condensation ratio obtained
was about 0.20 gH20/gD.G. It had been anticipated that a condensa-
tion ratio of 0.30 g H20/gD.G. could be achieved with the higher
gas temperature. The higher condensation ratio would further im-
prove particulate collection due to condensation effects. ^

PARTICULATE MASS CONCENTRATIONS AND SIZE DISTRIBUTIONS

The particulate mass concentrations were measured both with
total filters and cascade impacters. Tables 7-5 and 7-6 show the
measured particulate mass concentrations. The average particulate

-------
TABLE 7-1. F/C SCRUBBER SYSTEM OPERATING PARAMETERS FOR 36 SERIES

Run No.

Condenser How
T/s

Condensation Ratio
5 H20/q Dry Gas

Scrubber
cm W.C

36/1

47.3

0.13

36/2

47.3

0.21

36/3

47.3

0.18

36/4

47.3

0.06

36/5

47.3

0.23

36/6

47.3

0.19

36/7

47.3

0.14

36/8

25.2

0.03

36/9

25.2

0.12

36/10

25.2

0.08

36/11

25.2

0.10

36/12

25.2

0.10

36/13

25.2

0.16

36/14

47.3

0.17

36/15

47.3

0.17

36/16

25.2

0.13

36/17

44.2

0.15

36/18

47.3

0.03

36/19

47.3

0.12

36/20

47.3

0.05

36/21

47.3

0.18

36/22

25.2

0.11

-------
TABLE 7-2. F/C SCRUBBER SYSTEM OPERATING PARAMETERS FOR 37 SERIES

Run No.

Condenser How

1/s

Condensation Ratio
q H20/q Dry Gas

Scrubbe
cm W.

37/1

0.15

37/2

0.11

37/3

0.17

37/4

0.06

37/5

0.09

37/6

0.05

37/7

0.02

37/8

0.16

37/9

0.09

37/10

0.14

37/11

0.10

37/12

0.04

37/13

0.04

37/14

0.10

37/15

0.03

37/16

0.03

37/17

0.02

37/18

o.oo —

37/19

0.11

37/20

0.05

37/21

0.01

37/22

0.11

37/23

0.16

37/24

0.19

37/25

0.12

37/26

0.02

-------
TABLE 7-3. GAS FLOW

RATES, TEMPERATURES

AND MEASURED CONDENSATION RATIO FOR 36 SERIES

Gas Flow Rate

Sat. Inlet

Cond. Inlet

Cond. Outlet

Cond. Ratio

Run No.

DNm'/s

Temp °C

g H20/g D.6.
0.13

36/1

7.88

566

36/2

7.13

855

0.21

36/3

7.13

741

0.18

36/4

7.13

700

0.06

36/5

7.13

766

0.23

36/6

6.80

713

0.19

36/7

6.80

733

0.14

36/8

6.80

585

0.03

36/9

6.56

780

0.12

36/10

6.56

766

0.08

36/11

6.56

627

0.10

36/12

6.42

858

0.10

36/13

6.42

749

0.16

36/14

6.18

583

0.17

36/15

6.18

619

0.17

36/16

7.97

649

0.13

36/17

*6.00

792

0.15

36/18

7.00

460 I
683 •

0.03

36/19

6.60

0.12

36/20

6.08

605

0.05

36/21

5.90

648

0.18

36/22

5.94

ieui'

639

0.11

-------
Run No.

TABLE 7-4." GAS FLOW RATES,

Gas Flow Rate
ONmVs .

TEMPERATURES AND MEASURED CONDENSATION RATIO FOR 37 SERIES

Sat. Inlet Cond. Inlet Cond. Outlet Cond. Ratio
Temp °C Temp °C Temp °C i H.O/a D.G.

37/1

" 6.09

645

0H5

37/2

7.27

649

o.n

37/3

6.98

705

0.17

37/4

7.22

574

0.06

37/5

7.74

719

0.09

37/6

6.61

500

0.05

37/7

7.22

452

0.02

37/8

7.98

836

0.16

37/9

8.11

498

0.09

37/10

6.18

802

0.14

37/11

5.14

816

0.10

37/12

5.43

840

0.04

37/13

5.80

714

0.04

37/14

7.13

662

0.10

37/15

6.42

649

0.03

37/16

6.42

621

0.03

37/17

5.14

671

0.02

37/18

5.14

688

0.00

37/19

5.14

496

0.11

37/20

6.89

483

0.05

37/21

6.61

219 '

0.01

37/22

5.68

796

77 ,

0.11

37/23

5.80

740

0.16

37/24

5.00

752

0.19

37/25

6.50

690

0.12

37/26

5.64

610 |

0.02

-------
TABLE 7-5. PARTICULATE MASS CONCENTRATIONS

MASS CONCENTRATIONS, g/DNm3

Run No, Cond. Inlet Scrubber Inlet Scrubber Outlet

36/1 1.27 - 0.40

36/2 4.32 - 0.72

36/3 - - 0.50

36/4 2.50 - 0.52

36/5 2.63 - 0.73

36/6 2.57 1.13 0.47

36/7 - 0.77 0.57

36/8 1.00 0.85 0.41

36/9 1.51 1.33 0.40

36/10 T.07 0.77 0.45

36/11 - 0.73 0.45

36/12 1.58 - 0.51

36/13 1.17 - , 0.37

36/14 4.84 " - 0.15

36/15 1.56 - 0.71

36/16 2.74 1.19 0.73

36/17 2.77 - . 0.77

36/18 2.15 - 0.54

36/19 2.02 0.97 . 0.34

36/20 5.16 1.29 * 0.65

36/21 3.23 - 0.31

36/22 1.82 - 0.46

fifVt Z I. Oi>

-------
TABLE 7-6. PARTICULATE MASS CONCENTRATIONS

MASS CONCENTRATIONS, g/DNm*

Cond. Inlet Scrubber Inlet Scrubber Outlet

1.53 - 0.11

2.13 - 0.49

2.51 2.03 0.62

1.22 - 0.45
1.20 - 0.28
2.34 0.76 0.50
1.55 0.62 0.35
1.75 - 0.51
3.43 - 0.26

1.86 - 0.53
3.84 - 0.57
0.89 - 0.45

1.87 - 0.50

0.62

0.82 - 0.25

1.42 - 0.54

4.88 - 1.87

0.56 1.53 0.30

1.97 - 0.42

1.62 - * 0.42

3.10 - 0.40
2.40

1.94 - 0.54

5.62 - ' 0.70

7.23 1.72 0.77
1.74 - 0.60

¦zV&

(. ¦*>>

-------
mass concentration at the inlet to the condenser was found to be
2.4 g/DNm5. The design basis mass concentration was 1.8 g/DNms,
the average mass concentration was therefore found to be 33*. higher
than the design basis. , , „

The mass concentration varied significantly from run to rim
which reflects sampling during various periods in the charging
cycle. Figure 7-1 is a histogram showing the variation in the
measured mass concentrations at the inlet to the condenser. The
results of the first several preliminary tests showed that insigni-
ficant mass was caught in the impingers. This is expected because
of the nature of the source. As a consequence of these results
the impinger catch was not routinely analyzed for mass, only for
volume of condensed water.

The measured particle size distributions are shown in
Appendix "A" for the runs in the series 36 and 37. These results
also show variation from run to run. A statistical analysis was
performed to obtain an average condenser inlet size distribution
for use in determining the grade penetration curves. The averaging
calculations were performed numerically using a computer program
developed by Dr. Leslie Sparks of the U.S. Environmental Protection
Agency (Sparks 1979). Figure 7-2 shows the resultin-g-average con-
denser inlet size distribution. Figure 7-3 shows the cumulative
particle mass concentration versus particle aerodynamic diameter.

Figure 7-4 through 7 -49 show the measured penetration as a
function of particle size. The predicted penetration for the runs
are also shown in the dashed line. The agreement between predicted
and measured penetrations are good.

-------
0.30

e
o

N
«W

u
55

0.20

0.0

1.0 2.0 3.0 4.0 5.0 6.0

PARTICLE MASS CONCENTRATION, g/DNM®

Figure 7-1. Histogram of Condenser Inlet Mass Concentration

7.0

8.0

-------
10.0

CD
H

z
>*
o

OS
«
<

~ 5 10 20 30 40 50 60 70 80 90 95 98 99

MASS, i
-------
1.0 2.0 3.0 4.0 S.O

PARTICLE AERODYNAMIC DIAMETER, pmA

Figure 7-3.

Average Condenser Inlet Cumulative Size Distribution

-------
PARTICLE AERODYNAMIC DIAMETER, yraA
Figure 7-4. Predicted and Experimental Penetration Run 36/1

-------
V

u
u

99
98

SO
70

- 60
z
o

£ 50

w 30

0.3 0.4 O.S

1.0

2.0 3.0 4.0 5.0

10.0

dp, voA

Figure 7-5 Predicted and Experimental Penetration Run 36/2

-------
«->
c

f-.
V
p.

z
o

E-
w

2;
w
cu

99
98

95
90

80
70

60
50
40
30

2
1

0.3 0.4 0.5

1.0

2.0 3.0 4.0 5.0

10.0

Figure

dp, iimA

7-6 Predicted and Experimental Penetration Run 36/3

-------
4*)

c
€)

*gp»

as
w
a.

98 13

80
70

60
SO
40
30

2
1

0.3 0.4 O.S

1.0

2.0 3.0 4.0 S.O
dp, iubA

Figure 7-7 Predicted and Experimental Penetration Run 36/4

10.0

4t^ *

-------
0.3 0.4 0.5 1.0 2.0 3.0 4.0 S.O 10.0

dp, |imA

Figure 7-8 Predicted and Experimental Penetration Run 36/5

-------
0.3 0.4 0.5 1.0 2.0 3.0 4.0 5.0 10.0

PARTICLE AERODYNAMIC DIAMETER, pmA

Figure 7-9. Predicted and Experimental Penetration Run 36/6

-------
c
«>

o
u

V
p.

99
98

95
90

80
70

- 60

2S
O

- 50
£

w 30

20
10
5
2

0.3 0.4 0.5

1.0

2.0 3.0 4.0 5.0

10.0

dp, ymA

Figure 7-10 Predicted and Experimental Penetration Run 36/7

-------
<-»
e
*>
u

as
o

99
98

H ^O
ta

§ 30
&

2
1

0.3 0.4 O.S

1.0

2.0 * 3.0 4.0 S.O

10.0

dp, jimA

Figure 7-11 Predicted and Experimental Penetration Run 36/8

-------
iilt di.

-------
98

e 80

€>

£ 70

60
SO
40

0.3 0.4 O.S

1.0

2.0 • 3.0 4.0 5.0

10.0

dp, iubA

Figure 7-13| Predicted and Experimental Penetration Run 36/10

-------
0.3 0.4 0.5

10.0

2.0 3.0 4.0 5.0
dp, iimA

Figure 7-141 Predicted and Experimental Penetration Run 36/11

-------

£c

0.4 0.5

2.0 3.0 4.0 5.0

1C10

PARTICLE AERODYNAMIC DIA"ETER, jimA

Figure 7- 15, Predicted and Experimental Penetration Rim 36/12

?.!

-------
99
98

c 80
©

0 70

« 60
z

§ 50

£ 40

m 30

0.3 0.4 0.5

1.0

2.0 3.0 4.0 5.0
dp, iimA

10.0

Figure 7- 16¦ Predicted and Experimental Penetration Run 36/i3

-------
98

3 0.4 0.5 1.0 2.0 ¦ 3.0 4.0 5.0 10.

dp, ymA

re 7- 17 \ Predicted and Experimental Penetration Run 36/14

-------
+>
c

u
u
©
p.

•—I

0.3 0.4 0.5 „ 1.0 2.0 3.0 4.0 5.0 10.

dp, pmA

Figure 7-18 \ Predicted and Experimental Penetration Run 36/15

-------
98

.3 0.4 0.5

1.0

10,

2.0 3.0 4.0 5.0
dp, |imA

ire 7-19 ! Predicted and Experimental Penetration Run 34/K

-------
98

95
90

80
70

60
SO
40
30

2
1

0.3 0.4 0.5

1.0 2.0 3.0 4.0 S.O

dp, ymA

10.0

Figure 7-20 I Predicted and Experimental Penetration Run 36/17

-------
99
98

a 80

V
o

£ 70
o.

-r 60

2E»

m 30

0.3 0.4 O.S

1.0

10.0

2.0 3.0 4.0 S.O
dp, \mk

Figure 7-23 j Predicted and Experimental Penetration Run 36/20

-------
98

60
50
40
30

10.0

0.3 0.4 0.5 1.0 2.0 3.0 4.0 5.0

dp, jimA

Figure 7-22 I Predicted and Experimental Penetration Run 36/19

-------
f

I
r-

i
r

\
c

0.3 0.4 0.5

1.0

10.0

2.0 3.0 4.0 5.0
dp, pmA

Figure 7-21 ' Predicted and Experimental Penetration Run 36/18

C"/

J :

as
o

E-«

«
Z

99
98

80
70

60
SO
40
30

20
10
5
2

0.3 0.4 0.5

1.0

10.0

2.0 3.0 4.0 5.0
dp, ymA

Figure 7-24 j Predicted and Experimental Penetration Run 36/21

-------
99
98

- 60
sc

g 50

I 40

w 30
20

0.3 0.4 0.5

10.0

2.0 3.0 4.0 5.0

dp, iimA

Figure 7-25 1 Predicted and Experimental Penetration Run 36/22

-------
c

9>
U

-------
99
98

70 t
60

^ SO

S 40

P. 30

z 20
o

0.3 0.4 0.5

1.0 2.0 3.0 4.0 5.0

dp, ymA

Figure 7-27 ! Predicted and Experimental Penetration Run 37/2

-------
99

0.3 0.4 O.S

1.0

2.0 • 3.0 4.0 5.0
. dp, yi&A

7-28| Predicted and Experimental Penetrations Run 37/3

10.0

-------
0.3 0.4 0,5 1.0 2.0 . 3.0 4.0 5.0 10.0

dp, yAA

Figure 7-29! Predicted and Experimental Penetration Run 37/4

CfC

-------
99

iircridr-jjrd^

±:

-_l i U4 L_. ' L—l-..^i—^--L_4l4 i 4-,.,-L-L.

.3 0.4 0.5

1.0

2.0 . 3.0 4.0 5.0
dp, limA

r-30 ! Predicted and Experimental Penetration Run 37/5

10,

-------
I

99
98

e 80

£ 70

- 60
2 50

1 40

w 30
a,

0.3 0.4 0.5

1.0

2.0 3.0 4.0 5.0
dp, yi&A

Figure 7-31, Predicted and Experimental Penetration Run 37/6

10.0

-------
*5
o
u

l«

e
a.

s=
o

w
as

99
98

95
90

80
70

60
SO
40
30

2
1

0.3 0.4 0.S

1.0

2.0 3.0 4.0 5.0

10.0

dp, y®A

Figure 7-32 i Predicted and Experimental Penetration Run 37/7

-------
•M
C
«

w
a,

95
90

60
50
40
30

2
1

0.3 0.4 0.5

1.0

2.0 3.0 4.0 S.O
dp, y&A

Figure 7-33 , Predicted and Experimental Penetration Run 37/8

10.0

ac-

-------
o
u
u

se
o

E-

z
w
cx,

98 !=£

80
70

60
50
40
30

10
S
2

0.3 0.4 0.5

1.0

2.0 . 3.0 4.0 S.O

10.0

dp, jn&A

Figure 7-34f Predicted and Experimental Penetration Run 37/9

/ ^
r -

-------
©
u
N

99
98

8,0
70

2- 60

§ SO

§ 40

m 30
a*

20
10

0.3 0.4 0.5

1.0

2.0 . 3.0 4.0 5.0
dp, pmA

Figure 7-35; Predicted and Experimental Penetration Run 37/10

10.0

-------
e
v

35
O

CL)

99
98

60
50
40
30

'-—•j- "-"p ^|^—-j—" ~j~'

-i-i-i^T-i-i

E^.5r-=

in-.] - -ii.: ,^rz

J_I_L

0.3 0.4 0.5

1.0

2.0 . 3.0 4.0 5.0
dp, iiihA

Figure 7- 36 Predicted and Experimental Penetration Run 37/11

10.0

-------
~J

c
o

Vi
0)

z
o

E-

w
cu

99
98

r? SO

20
10

I ,1 ;

: '' I'

rr~

-T-t-

-j£r+

-4—L.

j-4:

¦Li..

!—i—t-t

:1r-

-j-r"f"—r---

——rH—

trm

* '4

—4-

Ei^iiE

¦ 4—Linr.

+- -«-r-

l—t i i-l

1—1—»—II

0.3 0.4 0.5

1.0

2.0 . 3.0 4.0 5.0
dp, pmA

Figure 7-37! Predicted and Experimental Penetration Run 37/12

10.0

-------
99

ire

3 0.4 0.5 1.0 2.0 . 3.0 4.0 5.0 10.

dp, viitiA

38 Predicted and Experimental Penetration Run 37/13

-------
w

99
98

-T 60

r? so

20
10

5
2

0.3 0.4 O.S

1.0

2.0 3.0 4.0 5.0
dp, \mA

Figure 7- 39 i Predicted and Experimental Penetration Run 37/15

10.0

-------
®

M
».

Illj"

-X^ in Ll-i

-X.

—1_

fe.rirrrjrr;: rrEcpr|rtr|:a:r^™(^
t r ~"tt~ "; !~~~j

»¦ | |i, »"¦¦ '¦»¦ y ii— ¦ f —in I i ¦ i *¦. > nw» ii w mi mi i m¦ i in | ¦ i ii i u i»| ¦ ti i j i i i ii i iiiii j ii' i i !¦!¦ »n—uiu Mamn—*4—

IMI»l^l^.i-lViy| --.^|- Mil ^Illii.m I.|| |I|..»|MH|||I|I T"'|' ¦—»""¦» ...I..I.MIII.MMIMA-. — -A.-. •.—¦4 » " *" ¦ — ^ — —- *• -—» * * ^ ^ )¦ "I'M-'»

0.3 0.4 0.5

1.0

2.0 . 3.0 4.0 S.O

dp, ymA

Figure 7-40 Predicted and Experimental Penetration Run 37/16

10.0

-------
0.3 0.4 0.5 1.0 2.0 - 3.0 4.0 5.0 10.0

dp, pmA

Figure 7- 41 \ Predicted and Experimental Penetration Run 37/17

^ ¦

-------
99

e 80

« 70
a

. 60

- 50

1 «

3 30

0.3 0.4 0.5

1.0

2.0 - 3.0 4.0 5.0

10.0

dp, piuA

Figure 7-42 (Predicted and Experimental Penetration Run 37/18

-------
u
N

- 60
ss
o

C 50
§ 40

n 30

ex.

20
10
5
2

0.3 0.4 0.5

1.0

2.0 3.0 4.0 5.0
dp, timA

Figure 7-43i Predicted and Experimental Penetration Run 37/19

10.0

ft -7

-------
©

N
©
A.

as
o

t-*

2
«
O.

99
98

~ 60

50
40
30

20
10

--^41.... ; nil-1, l^-S:

T".

i .: - r44 I

-+-+•

f=0E:

fc=t=3

0.3 0.4 0.S

1.0

2.0 3.0 4.0 S.O
dp, jimA

Figure 7-44! Predicted and Experimental Penetration Run 37/21

10.0

-------
©
u
u

* 60

E SO

§ 30

0.3 0.4 0.S

1.0

2.0 * 3.0 4.0 5.0

10.0

dp, ymA

Figure 7-45! Predicted and Experimental Penetration Run 37/23

s
I

I ' // /

-------
99

80
70

N
«
D.

- 60

n 30

20
10
5

0.3 0.4 0.S

1.0

2.0 3.0 4.0 S.O
dp, ymA

Figure 7-46 Predicted and Experimental Penetration Run 37/24

10.0

-------
99

• 60
z
o

E 50

0.3 0.4 0.5

1.0

2.0 3.0 4.0 S.O

10.0

dp, pmA

Figure 7-47: Predicted and Experimental Penetration Run 37/25

-------
i'-
*

99
98

«>
u
u
©
p.

« 60

2 so
5

P 40

B 30

20
10

0.3 0.4 0.5 1.0 2.0 * 3.0 4.0 5.0

* •
dp, vnA

Figure 7-48 ; Predicted and Experimental Penetration Run 37/26

10.0

-------
SECTION 8

ECONOMIC ANALYSIS

OPTIMUM SYSTEM DESIGN S

The optimum F/C scrubber system design for this application,
would not be significantly different(thaj* the demonstration plant
and the process design would remain substantially unchanged. The major
modifications recommended for an optimum system involve equipment
redesign and specification to reduce capital expense. Figure 8-1
shows a flow sheet for an optimum system.

The major area where redesign can provide significant cost
savings is a single combined condenser/scrubber vessel. The
diameter can be reduced from the size of the demonstration
plant by increasing the superficial velocity through the pack-
ing section. The depth of the packing can be reduced to
approximately 1 m, based on the results obtained in the demon-
stration plant.

The second item which can result in cost savings is an alter-
nate choice in cooling towers. The cooling tower chosen for the
demonstration plant was dictated by space requirements. The cool-
ing tower had to be elevated on a platform above the saturator and
sump. Consequently the most compact tower was chosen,which was
not the most economical or energy efficient.'

In the course of the project bids were obtained by other
manufacturers. A wood-filled cooling tower would be the
most likely choice if space were not at a premium. Such a j

tower would require only 30 HP, compared to 60 HP for the tower
used on the demonstration plant. The wood packing may have some
advantage over epoxy coated steel in terms of corrosion resistance.
A second circulating water pump would be required for the system
but the total cost would be less /sine?®: an expensive structural

-------
steel platform would not be needed.

The saturator design would be changed in an optimum system.
The saturator was designed to allow the equipment to be fit into
the available space at the foundry. An optimum system would have
a smaller, more compact saturator.

The settler used on the demonstration system provided accept-
able performance, however the condenser, scrubber and sump all were
subject to deposition of fine sandy material which required removal
at frequent intervals. The material which settled out was fairly

large, which is apparent from the fact that it settled out in these

process vessels. Sloped bottoms should be used in all of these
vessels.

It is also recommended that a moving belt filter be used in-
stead of the gravity settler to remove this sandy material. These
can be purchased at a lower cost than the settler which was used
on the demonstration plant.

CAPITAL COST ESTIMATES

The capital cost of an optimum F/C system would consist of
the total equipment costs and other direct costs such as installa-
tion, piping, electrical, etc., along with indirect costs such as
engineering, construction overhead, contingencies, etc. The total
equipment costs were calculated as described below. For equip-
ment identical or similar to that purchased for the demonstration
plant, costs were scaled up from those incurred for the purchased
component. For the condenser/scrubber vessel, cost estimates were
prepared with standard techniques and then compared to purchased
cost for the demonstration plant scrubber and condenser vessels.
The fan and motor costs were obtained from the data presented by
Neveril et al. (1978). Tables 8-1 and 8-2 show the total equipment
costs for the F/C system and high energy scubber respectively.

The other direct costs were computed by means of ratio
factors obtained from Peters (1968) which are based on equipment
costs. These factors were confirmed using cost data from the
Demonstration Plant. The indirect costs have been estimated in the

-------
r.-

' f

CAP
LIFTER

CUPOLA AFTER-
BURNER

PILOT
BURNER

CONDENSER
/SCRUBBER

SATURATOR

P-l

P-I Saturator Pump

P-2 Cooling Tower Pump

P-3 Condenser Pump

P-4 Water Treatment Pump

ft) EXHAUST
V J FAN

P-2

P-4

COOLING
TOWER

P-3

SOLIDS

/n&

SmJ-i

7 ir

Figure 8-1. Optimum F/C Scrubber System Flow Diagram

A-

£&€*t,

-------
TABLE 8-1. SUMMARY OF EQUIPMENT COSTS
F/C SCRUBBER SYSTEM

Saturator pump & motor
starter

Cooling loop pimps & motors
starters

Exhaust fan

motor & starter

Cooling tower
fan starter

Saturator shell
Internals

Condenser/scrubber shell
Internals

Tank

Belt filter

Unit Cost

$ 2,290
240

6,950
580

3,890
8,290

29,370
280

3,650
1,300 "

20,000
8,000 ,

4,630

12,900

TOTAL EQUIPMENT COST

-------
TABLE 8-2. SUMMARY OF EQUIPMENT COSTS
CONVENTIONAL SCRUBBER

Unit Cost Total Cost

Saturator pump & motor $ 2,290

starter 240 $ 2,530

Scrubber pump S motor 3,680

starter 195 4,075

Saturator vessel 3,650

internals 1,300 4,960

Exhaust fan 5,350

motor & starter 18,450 23,800

Scrubber vessel 14,600

internals 10,900 25,500

Belt filter 12,900 12,900

TOTAL EQUIPMENT COST $73,765

-------
same manner with factors suggested by Peters (1968). Tables 8-3
and 8-4 show the indirect costs for the F/C system and the high
energy scrubber. All costs were adjusted to December 1979 by
means of the M§S cost index.

OPERATING COSTS

The operating costs for the air pollution control system
consist of the annual cost of the utilities (power and water),
raw material and maintenance. Table 8-5 shows the power require-
ment for both the F/C scrubber system and a. conventional scrubber.
The F/C scrubber system is shown to require only 651 of the power
of the conventional scrubber. The exhaust fan for the conventional
scrubber requires more than twice the power of the F/C system.

The total operating costs are summarixed in Table 8-6. The
power cost was estimated at 6.5#/Kwhr. The water cost estimate
was 3.5t/l,000*. The raw materials included soda ash for neutra-
lizing the scrubber liquor at 26.5f/Kg and flocculant to improve
water treatment at $1.15/1. The total operating cost for the F/C
scrubber system was found to be J-frfTSM compared to $*-H-75TT5\for
the conventional scrubber. Xhe annua? operating cost of tire F/C
scrubber system was about /St/of thy cost for a conventional scrubber
for this application

/it- e'U

% n*™dU, 1 ^

Vi 1 / 0

-------
Direct

TABLE 8-3. DIRECT AND INDIRECT COST
F/C SYSTEM

Equipment

Installation
Instruments
Piping and Ducting
Electrical
Site Preparation

Total Direct Cost

Ratio

1.00/

0.40

0.10-'

0.40'

0.107

0.05

2.05;

Cost, $

$102,380
40,952
10,238
40,952
10,238
5,119

$209,879

Indirect

Engineering
Construction Overhead
Contractors Fee
Contingency

Total Indirect Costs
Total Capital Investment

0.40 '
0.45 <
0.10 ^
0.40 ^

1.35 '
3.40 7

40,952
46,071
10,238
40,952

$138,213
$348,092

U'r. /!- '

k ^ k v

M »

&1L

-------
9

Direct

Equipment

Installation

Instruments

Piping and Ducting
Electrical
Site Preparation

Total Direct Cost

TABLE 8-4. DIRECT AND INDIRECT COST
CONVENTIONAL SCRUBBER

Ratio

1.00
0.40
0.10
0.40
0.10
0.05

2.05

Cost, S

$ 73,765
29,506

7,377
29,506
7,377
3,687

$151,218

Indirect

Engineering
Construction Overhead
Contractors Fee
Contingency

Total Indirect Costs

0.40

0.45
0.10
0.40

1.35

29,506
33,194
7,377
29,506

$ 99,182

Total Capital Investment

3.40

$250,800

-------
TABLE 8-5. SUMMARY OF POWER REQUIREMENTS

Power, Kw (HP)

Exhaust fan
Saturator pump
Scrubber pump

Cooling water pimps
Cool frig tower fan

F/C Scrubber
156 (210)
10 (13)
2 (2)
33 (44)
22 (30)

Conventional Scrubber
332 (445)
10 (13)
3 (4)

Total Power Required

223 (299)

345 (462)

TABLE 8-6. SUMMARY OF ANNUAL OPERATING COSTS

Item

Capital Cost
Maintenance

Power Cost *
Water Use
Raw Materials

Unit Cost

ft ///v Jfifi ft tCw <;

10 yr, S.t. depreciation
Ut>

6S of e^espfflebt cost

6.5
-------
REFERENCES

Calvert, S., J. Goldshmid, D. Leith, N. Jhaveri. Feasibility
of Flux Force/Condensation Scrubbing for Fine Particle
Collection. A.P.T., Inc. EPA Contract No. 68-02-0256. "
NTIS #PB 227 307. October 1973.

Calvert, S., J. Goldshmid, D. Leith, N. Jhaveri. Scrubber
Handbook. A.P.T., Inc. EPA Contract No. CPA-70-9S.

NTIS #PB 213 016. August 1972.

Calvert, S., S. Gandhi. Fine Particle Collection By A Flux-
Force/Condensation Scrubber: Pilot Demonstration.
A.P.T., Inc. EPA Contract No. 68-02-1869. NTIS #EPA-
600/2-77-238. December 1977.

Calvert, S., N. Jhaveri, T. Huisking. Study of Flux Force/

Condensation Scrubbing of Fine Particles. A.P.T., Inc.
EPA Contract No. 68-02-1082. NTIS # EPA-600/2-75-018
August 1975.

Green, H.L., W. R. Lane, Particulate Clouds £. I F. N. Spon
Ltd. London, 1964.

Neveril, R.B., J. V. Price and K. L. Engdahl, Capital and Operat-
ing Costs of Selected Air Pollution Control Systems Parts
II and IV. J. Air Pollution Control Association V28 N9

P963, September 1978 and V28 N10, October 197S.

McCain, J.D., Clinard, G.I., Felix L. G., Johnson J. W., A

Data Reduction System for Cascade Impactors. EPA No. 600/

7-78-132a, July 1978.

Peters, M.S., K. D. Timmerhaus, Plant Design and Economics for
Chemical Engineers, McGraw-Hill Book Company, 1968.

Yung, S., S. Calvert, H. Barbarika, Ventrui Scrubber Performance
Model. Environmental Science and Technology V 12 p.456,

April 1978.

-------