Nitrate Removal from Contaminated Water
Supplies. Volume 1. Design and Initial
Performance of a Nitrate Removal Plant
MoFarland Mutual Water Co., CA
Prepared for
Environmental Protection Agency, Cincinnati, OH
Jan 87
PB87-14S470
SBHBSBS
U.S.
-------
PU87- U5U70
EPA/600/2-86/115
January 1987
NITRA1E REMOVAL
FROM CONTAMINAfCD WATER SUPPLIES
VOLUME I. DESIGN AND INITIAL PERFORMANCE OF A
NI1RATE REMOVAL PLANT
by
Gerald A. Guter
Boyle Engineering Corporation
Bakersficld, California 93302-0670
Cooperative Agreement No. CR808902-02-0
Project Officer
Richard Laucu
Drinking Water Research Division
Water Engineering Research Laboratory
Cincinnati, Ohio 45268
WATER ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OHIO 45268
-------
TECHNICAL REPORT DATA
(Plrcse ‘r J lnjtr.jcl Ion Ilu ,r,rl’ , b Jcue 10’PprI:nt1
I REPORT NO
EPA/600/2-86/115
3 NE CiV.LN1 S ACCLS ..O’ .O
1 4 47 Oi s
4 TITLE AND 5U9flT
Nitrate Removal from Contamin3ted Water Supplies
Volt e I, Design and Initial Performance of a Nitrate
Re. ova1 Plant
5 R(P$ RI D4 ( —______
uanuory 987
-
6 P(OMMINGOMOANIZA TION CODE
7 AUTI4ORISI
Gerald A. Cuter
PEORMINC ORGANI2A IO . NIPONT NO
9 PEAFORMIN5 ORGANIZATION NAME AND ADOR SS lOPc)C jirj j NO
Boyle Engineering Corporation
Bakersfield, CA 93302—0670 For H CONY ACT,GRANT N
McFarland Mutual Water Co., 606 Second Street,
CR808902
McFarland, CA 93250-1118
I? SPONSORING AGEN y NAME AND ADDRESS 13 TyPLO REPOIIT AND PENIOD COVERED
Water Engineering Research Laboratory—Cincinnati, OH fltCiiPrJt_Reperi .j98l-8 .
Office of Research and Development ‘ SPONSORING . CEN:. coot
U.S. Environmental Protection Agency
Cincinnati, OH 1 5268 EPA/60O/1
15 SUPPLEMENISRY NOIES
Project Officer: Richard P. Lauch
lb MU TNACT
This report reviews the de.lgn. construction, and operation of a 1—a.gd nitrate removal plant
in McFarland, CA. The pl.nt treats veil water for dnmestic use. Nitratee are reduced tr approil—
mateiy 15.8 mg U0 3 —N/L to well below the maximum contaminant level of 10 og ? 10 3 —NIL.
Continuous daily (24—hr.) operation of the plant was made possible by utom tic operation.
Automatic nitrate monitoring of product water was perfot-ned once en hour through the use of eodifIe i
ion chromatography. Daily records of (lotis, water quality, electrical consumption, salt usage. and
manhours were kept to determine operating costa.
The total vastewater produced by the nitrate plant was 3.39% of the amount of water delivered
to the distribution system from the well. The treated water was 751 of water delivered. Saturated
brine was 0.09 . dilute brine was 0.49%, rinse water was l.7& , end backwash water use 1.14%. L i i
percentages were of the blended water delivered to the distribution cyatem. all waste from the
plant was discharged to the McFarland municipal vastowater treatment system. vith ultimate discharge
to 120 acres of cotton and alfalfa crops.
The aaount of water treated by each ion exchange veo el before regeneration was 165,900 gal.
(260 bed volumes). The amount of salt used per regeneration was 6.35 lb/ft 3 of resin.
Capitol coats totaled $35,638 for a 5—ft. bed system. Orerat ion end maintenance costa were
$0.13 per thousand gal. when the system wee operating at 1 tsgd. Total coats, including operatlone
and maintenance (O4M) and amortiLed capital, were $0.25 per 1000 gal. v4 en operating at design
capacity of 1 vgd.
I DESCRIPTORS
bIOENTIPIERS’OPCN INDED TERMS
7 tv WORDS AND DOCUMENT ANALySIS
IS DISTRIBUTION STATEMENT
Release to public
IS SFCURUY CLASS iri. Nrp(I t)
Unclassified
SECURITY CLASSITII ,Jpa 1 ,
Unclassified
11 ‘ U 1 U 4 jAL.LS
??PRIC(
C COSATI I eli Groap
EPA Fo. , 2220—1 (5.. 4..77) PRC eOuS coy.o ,. $ O IOLCTE
-------
DISCLAIMER
The information in this document has been funded wholly or in
part by the United States Environmental Protectio.-i agency under
assistance agreement number CR 808902—02—0 to McFarland Mutual
Water Company. It has been subject to the Agency’s peer and
administrative review, and it has been approved for pubUcation
as a EPA document. Mention of trade name” or commercial
products does not constitute endorsement or recommendation for
use.
U
-------
FOR1 WORD
The U.S. Environ entaj Protection Agency is charged cy Congress
with protecting tne Nation’s land, air, and water systems.
Under a mandate of national environInent .1 laws, the agency
strives to formulate and implement act ions leading to i
compatible balance between humdn act1vi i , and the ability of
natural systems to support and nurture life. The Clean water
Act, the Safe Drinking Water Act, and the Toxics ubsnc
Control Act are three of the major congressional laws that
provide the framework for restoring and maintaining the
integrity of our Nation’s water, for preserving and enhancing
the water we drinK, and for protecting the environment from
toxic substances. These laws direct tne EPA to perform
research to define our environmental problems, measure the
impacts, and search for solutions.
The Water Engineering Research Laboratory is that L mponent of
EPA’s Research and Development program cOncerned with
preventing, treating, and managing municipal and Lnduz trial
wastewater discharges; establishing pract ccs to control, and
remove contaminants from drinting water and to prevent its
deterioration during storage and distribution; and assessin.J
the nature and controllability of releases of toxic SuOStCnCeS
to the air, water, and land from nanufacturing processes arid
Subsequent product uses. This publication is one of the
products of that research and provides a vital communication
link between the research and the user community.
The pollution of our Nations’ groundwater has been called the
environmental problem of the l980s. When polluted grounduate;
serves as a source of public drinking water, pollutants must be
removed to levels below standards regulated by the Sate
DrLnking Water Act (Puolic Law 93—523). Nitrate is one of the
pollutants frequently found in groundwater used for drinking
water supplies and is the cause of serious and Occasionally
fatal poisonings in infants. This report summarizes studies
being conducted in behalf of small communities on processes Lor
removal of nitrate from groundwater supplies.
•l1 l
-------
ABSTRACT
ThLS report reviews the de 1ga, Construction, and operation of a
!—igd nitrate removal plan, in HLFarland, California. The plant
treats groundwater pumped from one of the we! Is supplying water
tor dume tic use. Nitrates are reduced trom approximately 15.8
mg/L t ’ 0J-h to well below the ciaxi ’num contaminant level of 10 mg/L
NLJ3—N. Included in the design cor.sidcrations are such factor’, s
water supply, health aod safety, level of technology, location,
cap tc ity, regenerahi on trequc ncv, water qua Itty, operati onal
‘ -equen e, brine disposal, automatic uper -ition, and petotminitce
monitoring. Itte procedares br both manual and automatic
Operation are discussed.
Continuou 5 daily (24—hr.) opetation of the plant was m 1 de
poscible by automatic operation. Tue presence of the operator is
required for approximately 1 hr. per day to check performance.
u tome tic iii trite men i tort ‘ig Ut product water was port ormed once
on hour through the use of mod iii d ion chromatography. Daily
records of flow’,, w.,tc ’r quality, electrical conaunption, salt
usage, and manhours were kept to uetermtn, operating .osts.
Th t tal wastewater produced by the nitrate plant was 3.i97 of
the as. ‘tint of water delivered to the dist ribut Ion system [ rein the
well. The treated water was 75 of water delivered. Saturated
brine i. 0.09, dilute brine was Q,4( , rinse wa ei was 1.75%
and b ,-ic a ’ h water was 1.l4Z. All percentages were of the
hl ded ‘. her deliecreti to th di iribu i,,n s’,stem. All waste
from thc ; ant w is d ischarged to the McF. -ir [ and munic i pal
w ,IsleWalt ’r rearaertL system, with ultimate discharge to 120 acres
ot cotton . id alfalfa crops.
The amount - water treated by each ion exchange vessel bafore
regeneratlor. was 165,900 gal. (260 bed volumes), ‘the amount of
salt used pe regeneration was 6.33 lb/ft3 of reci.n.
Lapital coi c totalea $3 11,I IS for a 3—ft. bed svsteni, and
355,63 (or - 5—ft. hed s\stcm. Operation and maintenance Costs
were 0. 13 pe - thousand gal. when the sybtem was operating at I
i gd. Total c ’3ts, in lading operations and maintenance (0&M) and
dcnortLLed cap. ral, were $0.25 per 1(00 gal. whsn operating at
design capacity of 1 mgd.
This report w submitted in fulfillment of ( ooperatLve Agreement
No. CR8U8902—0!—0 by NcFarland Mutu.il Water Company under the
sponsorship ot the U.S. Environmental Protection Agency. Boyle
l ngineering Co jioratlon served as subcontractor. hits report
covers the per id June 1, 1984 to November 30, 1984, and work was
completed as o. October 30, 198 .
iv
-------
TABLE OF CON1 LNTS
DisLl.i bier
loreword .
Abstr aLt
lgurcs
Fables
‘u knowicugments
2
5
• S
• 17
• 21
24
30
33
35
35
37
37
46
52
55
62
62
65
78
78
78
78
93
95
95
107
118
124
References 130
ii
iLl
lv
vi
\,iii
1. Introduction
2. COnLIUSIOUS and Recommendations
3. Construction of Plant
Background of Project
Design and Operating Parameters
}lcadworks
Ion Lxchangc Vssels
Vessel Intcrn ils
Brine System
Product Delivery System
l’rocess Control
Semi—automatic Operation
AuLomati . Operation
4. P [ ant Operrilicn
Check Out and Start Up
I)eclas’,rfication Tests and Nanu il Operation
Analysis of I niLial Operation
Semi—automatic Operation
Automatic Operation
Operational Problems
5. Plant Performance
Daily Operating Records
Nonthly Reports
Chemical Composition of Well and Product Water
Evaluation of Primary Plant Performance
Criteria of Primary Performance
Estimates of Salt Dosage, Brine Use Factor,
and Nitrate Leakage
Effects of Underusing a Led on Brine Efficiency
Secondary I’lani Performance Factors
6. Cost Analyses
V
-------
Typical Nitrate Leakage Dependence
on Salt Dose
Typical Break Through Curve
Flow Diagram
General View of Plant
Pump Discharge Piping
Piping to Distribution System
Discharge from Booster Pump .
Motor Operated Blending Valve
General View of Headworks Piping
Front of Ion Exchange Vessels
Vessel Piping - Left Side
Vessel Piping — Right Side .
Rear View of Ion Exchange Vessels
Internals of Ion Exchange Vessels
Brine System Piping
Batch Reset Meters
Brine and Makeup Flow Meters .
Treated, B3ckwash, and Rinse Water
Main Control Panel
Main Control Panel Interior
Ion Chromatograoh
Laboratory and Control Building
Interior West Side
Laboratory and Control Building
Interior East Side
Run No. 01114-P3
Run No. 12143—Pi.
Daily Report
Manual Report
Regeneration Report
Operating Variables Print Out
Alarm Report: Power Failure .
Alarm Report High Conductivity
Nitrate Level Recording
Electrical Conductivity of Raw
and treated Water
Continuous Seven Day Reocrd of Flow
Rate to Distribution System
Daily Operating Data rormate No. 1
Pace
10
13
16
19
23
25
26
27
28
29
31
32
33
34
36
37.1
38
Flow Meters 39
41
42
43
44
45
53
54
66
67
68
69
70
71
72
73
74
79
FIGURES
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
Preceding page blank
vii
-------
‘).iily Op . .ating Data Format No. 2
Daily Operating Data Format No. 3
DaLly Record o Meter Reading Format
Daily Pressure Readings Format
Chemistry Report Format
Water and Column Chemistry
January 198.1 (100%) .
42 Water and Column Chemistry
November 1984 (100%)
43 Water and Column Chemistry
June 1984 (260 DV) .
44 Water and Column Chemistry
July 1984 (260 DV) .
45 Water and Column Chemistry
August 1984 (260 DV)
46 Water and Column Chemistry
September 1984 (260 DV)
47 Water and Column Chemistry
October 1984 (260 DV)
48 Water and Column Chemistry
November 1984 (260 DV)
49 Water and Column Chemistry
June 1984 (100%)
50 Water and Colum’. Chemistry
July 1984 (100%)
51 Water and Column Chemistry
August 1984 (l00) .
52 Water and Column Chemistry
September 1984 (100%)
53 Water and Column Chemistry
October 1984 (100%)
54 Water and Column Chemistry
November 1984 (100%)
55 Comparison of Plant Data with Computer
Generated Effluent Histories
56 Input Data for Computer Curve
of Figure 55
57 Waste Water Conductivity History
Page
80
81
82
83
84
96
98
101
102
103
• . 109
110
113
117
118
121
N umbc r
FIGURES
36
37
38
39
.20
41
104
105
106
108
111
112
Yii ’
-------
TABLES
Number Page
1 Composition of McFarland Well Water
(ppm) in 1980 6
2 Estimated No. of Regenerations per Vessel
per Day 11
3 Bed Characteristics and Target Flows 18
4 Resin Purchases 46
5 Wash Water Composition 49
6 Position of Sample Ports 50
7 TDS of Wash Water 50
8 Chemical Analyses for Initial Runs 52
9 Data for Run 0114—P3 57
10 Data for Run ].2143—P1 58
11 Effectiveness of Declassification 59
Comparison of Actual and Calculated Performance 60
13 Regenerant Efficiency - Plant Data 61
14 Plant Records June 1984 85
15 Plant Records July 1984 86
16 Plant Records August 1984 87
17 Plant Records September 1984 88
18 Plant Records October 1984 89
19 Plant Records November 1984 90
20 Summary of Monthly Data 91
21 Changes in Well Water Composition Over 11 Months 92
22 Monthly Anion Analysis Certified Laboratory . . . 94
23 Monthly Brine Efficiencies 99
24 Summary Comparison of Actual Chemical Data
with Predicted 114
25 Estimates of BUF for Full Bed Use and Partial
Bed Use 115
26 Secondary T lant Performance Factors 120
27 Capital Costs, McFarland (1983) 125
28 Cost Detail for Equipment and Construction . . . 126
29 Operation and Maintenance Costs 129
30 Total Cost of Plant and comparison to Present
Water Cost in McFarland 129
31 Present Costs of Water in McFarland 129
ix
-------
AC K N O LED G CM C NT S
The author wishes to acknowledge the professional assi itance
and encouragement of the following:
Mr. David L. Hardan, P.C., Vice Preuident,
Boyle Engineering Corporation.
Mr. Ernest Kartinen, P.C., the program manager of this
project for Boyle Engircering Corporation.
Mr. Steven PaUska, P.C., Mechanic3l design engineer,
Boyle Engineering Corporation.
Mr. Dick Lauch, the prolect officer for CPA.
A special note of grat itude is due to the i3oard of Directors of
McFarland Mutual Water Company and to Mr. C rrol Hurst,
Manager, whose suppnrt of this project is much appr* ciated.
x
-------
SECTION 1
INTRODUCTION
This report reviews theoparationofa 1—mgd nitrate removal plant at
McFarland, California. The plant and supporting equipment are
described, andananalysisof thecapital costof construction and
theoperation and m jntenance (O&M) costs arealsopresented. The
dataonwhichthisreport is baseriwereobtainedduring the initial
ad)ustment period of the plant and during th first 6 months of
automatic operation ending November 30, 1984.
T’ e plant uses the ion exchange processwi.thcommercia llyavai lablo
resin. The process design is based on the research and pilot
tudtes performed under a previous U.S. Environmental Protection
Agency (EPA) coopera.tive agreement. Reference 1. The design and
opPratton of the plantwcre supported byMcFarland ’utualWatarCo.
(McrMWCo) and EPA under 000peratLve agreemtnt Nos. CR60802—OlO and
CR8U8d2—020. Construction of the plant was mad possible w th
funds from MC.M iCo, The Kern Co. Conmunity Development Agency (CDA)
nd the K. rn County Water Agency (KCWA) . The CDA is funded by the
U.s. Department of Houaintj and Urba. Devclopm nt (IIUD)
Tnis report is the first of a two volume final, report under the
e lsting grant and is restricted to the general ubjmct of the
initial operation of the plant. The second volume will include a
report on the continuing operation ot tie plant for several
additional months.
-------
SECTION 2
CO JCLuSzo s M4D RECOMME 40ATIONS
1. The plant was automatically operated for a 6 month period and
gave the following performance characteristics averaged over
the operating period:
a. Nitrdte leakages averaged 5.2 mg N03—N/L or 23.2 mg in a
blend of treated and untreated water.
b. The blend consisted of 76.1% treated water and 23.9%
untreated water.
c.
2.49 lb per thousand gal of blended water.
d.
Per equivalent of flL trate removed and var ied f rom a low of
8.3 to a high of 11.8.
e. Water recovery was 96.7% of the water pumped. The
remaining 3.3% was discarded as waste brine and wash
water.
f. Wastewater per thousand galot blended water consjstedof
0.92 gal of saturated brine (4.9 gal of dilute brine) 17.6
qal of rinse water, and 11.4 gal of backwash water.
2. Maximum automation was used Successfully to satisfy the
minima’ manpower reQuirements of a small water system
operator. The plant was designed and is being demonstrated
primarily with the needs of small communities in mind where
wells and distribution systems are airead! in place. The
plant operates atawell site rather thanasacentral treatment
plant.
3. Raw water composition varied during this period of operation.
Nitrate varied from 16.0 rig N03-N/L or 71.0mg N03/L to ll.O7mg
N03—N/L or 49.0. Sulfate varied from 115 to 60 mg/I. This
pr vlded the opportur ty to measure the effect of changing
waLer con ositjon a. plant performance.
4. Resin beds were operated at 76% capacity during this initial
adjustment period to prevent over—runs which could occur
because of operational problems.
2
-------
5. The effect of operating at less than 100% capacity is estimated
to be a decrease in brine efficiency of approximately 18%.
. Brine efficiency, nitrate leakage, and bed volumes to nitrate
breakthrough can be accurately predicted from ion exchange
theory. Computer based programs bei.ng developed can sneulate
effluent histories and are comparable to those obtained from
the plant. They also give chromatographic distributions of
ions within spent beds.
7. A 3—ft. resin bed depth was used during this period of
operation. A S—ft bed nil be used in future tests to obtain
comparative data.
8. The power consumed by the plant is 244 kwh per million gal of
blended water. This amount is 10% of the total power required
for pumping at the well site.
9. Capttal costs were $311,118 (or a plant wtth the 3—ft—deep
resin bed and $355,683 for a plant with a S—ft bed. The total
costs are .24S per 1000 gal of blended water for the S—ft bed
plant (1983 costs). During this report period the plant was
operated at only 13.7%. capacity. The overall cost to the
H oFar land co smuri ity far nitr ate removal during this per iod was
$0.l62 per 1000 gal water consuned.
10. The plant is totally auto’natic in operation with automatic
nitrate analysis for monitoring, and automatic shut down if
nitrate exceeds the MC I . in the product .tater. Computer
printouts of operating data are obtained on a daily basis and if
alarms occur.
11. Operator tasks are reduced to approximatly 1 hr per day and
include routine inspection, maintenance, and record-keeping.
12. Nitrate renovai is economically and technically feasible by
the ion exchange process. The most undesirable feature is the
production and disposal of waste brine. At McFarland during
this report period, approximately 1300 lbs of waste salts were
disposed of in the plant wastewatet daily by discharging to the
municipal waste water system. L U the plant were operated 24
hrs/per day, the daily salt discharge would be 2500 lb in 33,000
gal of wastewater. Close monitoring of soil and plant
conditions at the disposal site is being conducted.
13. Although nitrate removal by the ion exchange process is largely
beingconsideredasaprocessadaptable for smalicommunities,
it is the latter who will find the waste disposal problems the
most difficult to solve. Improvements in the process are
still required toreducequantitiesofwastesalts. These can
3
-------
probably be accomplished by use of highly selective nitrate
resins, br me recirculat Ion, recovery and separation of sodium
nitrate and sodium chloride, and close adjustment of plant
operation to changes in raw water composition.
14. Plant shut—downs were due to malfunctions of electrical and
mechanical equipment and leaks in plastic pipe. All repairs
were handled by water company personnel.
15. The adjustment and operation of the plant was complex because
by using the same microprocessor for plant control and dat3
collection and reporting. Considerable operating time was
lost due towriting and testing thedatacollectionportionof
the program. The controller required prograsnning by ladder
logic which is cumbersome as a computer language. A separate
computer is recommended for data collection at a similar
installation.
16. Ion chromatography is satisfactory for routine anionanalysis
and research but definitely requires improvement for
continuous on—stream plant monitoring. Additional research
onacontinuousnitrate monitorthatwillshuttheplantdown
if nitrate concentration exceeds the MSCL is recommended.
17. Further development of the nitrate selective resin is
recommended because use of a nitrate selective resin would
eliminate the possibility of nitrate dumping.
18. Further research on waste brine disposal and brine reuse is
recommended toelimiate the problem of waste nitrate and waste
salt contamination of the disposal area and u derLying
groundwater.
4
-------
SECTION 3
CONSTRUCTION OF PLANT
BACKGROUND OF PROJECT
McFarland
McFarland, California is an incorporated city of approximately
5,000 persons (4,177—1970 census arid 5,765— 1980 census) located on
U.S. Highway 99, approximately 20 miles north of the city of
Bakersfi rld in Kern County. Like many small San Joaquin Valley
co’nriunit ies, muchofMcFarland ’SeCOflOmyiSbaSedoflagri cu ltureand
agricultural—related business.
Water Supply
TheMcFarlandMutUalWater Companysupplieswater from itswellsand
through its disLrtbutiori system for municipal use. Theonlywater
source at present i underlyingcjroundwater. Sixwells are located
in McFarland. Use of Well No. 3 has been discontinued for public use
due to high nitrate levels. Well 2 is the locdt ion of the nitrate
plant now in operation. Recent analysis of Wells 1 and 4 show
nitrates above the Maximum Contaminant Level.
Water quality of four wells containing high nitrates are given In
Table 1. Due to recent and projected development trends, two new
wellswere recentlyconstructed toserve the developing areas ma
remote part of the City and will shortly be in operation.
The static water level in the wells is approximately 250 feet below
the surface. Except for excessive nitrate levels, the water is of
generally good quality.
McFarland Nitr .te HistorY
Water quality data show significant nitrate levels present inwater
supplied by r icMWCo. before 1965. In 1978 McMWC0. requested and
received a Cooperative Agreement to study treatment from the U.S.
€PA. As a result of this work, as reported in Reference 1,
significant insight into the ion exchange process was achieved and
it was realized that operating cost reductions and waste brine
reduction could be anticipated through further research on nitrate
selective resins. However, due to the pressing need to have an
operat ing system McMWCO . dec ided to proceed with the c onst ction 0f
a full scale plant using readily available ion exchange technology.
The design allows adequate instrumentation metering and process
5
-------
TAGLE 1 CC 4P0 !T!ON 01 E L: ND WELL ATER (r,pi) IN 1930
Nell No .
Ite’n I 3 4
Date 5- 3-80 4-9-30 5-1-CO 4-16-30
Ca1c iu. 28 83 1 o 78
Sodium 50 65 100 72
Bic rbonate 83 102 121 95
Chioide 23 85 9 4 51
Sul 3te 51 1 35 310 132
Nitr atc+ C7 98 47
TOS 2 4Gb 827 485
ph 7.7 7.2 7.3 7.7
Analyses on 5/31/73 sho c ’1 sulfate levels of 261 pp.i and
nitrate levels of 73 ipii
÷Concentratlon of ntrat listed as n g N03/L.
To cor.vert to r.g N03-N/L divide by 4.43
6
-------
flexibility to allow close monitoring and testing of future process
improvements as well as provide detailed operating costs for a base
line or state—of—art system. ThedesignalsOallOWSUSeOf improved
resins and expansion for brine recirculatiOn and recovery.
Basis of Plant Design
The major considerations given to the design of the McFarland plant
can be listed as:
Health and Safety
Level of Technology
Location
Capac ity
Regeneration Frequency
Water Quality
Operational sequence
Brine Disposal
Automatic Operat ion
Monitoring of Performance
Health and Safety
This is the major consideration and took precedence where design
conflicts arose. The plant design was reviewed by the California
State DivisiOn of Health who issued an operating permit on May 13,
1983. Prior to issuance of the pernit a design review was
conducted. The major concern was that:
1. Brine be isolated from the brine water supply system. This is
accomplished with a double check valve.
2. Waste brine and wash water be isolated from the distribution
system. Accomplished by double valves or “block and bleed”
arrangement.
3. Nitrate levels in supply water be kept below 10.0mg N03-N/L or
44.3mg N03/L. and preferably at 6.8mg N03—N/L or 0.0 mg N03/L.
4. A Class 2 State certified operator be made responsible for
plant operation. Two employees of McMWC0. will qualify for
this certification.
The reason for the first two requirements was the concern of health
officals that waste nitrate in concentrated amounts could
contaminate the water supply if valves malfunctioned and cross
connections occured. This is a major concern of health otficials
for all water treatmentSyStemS. ionexchangetsbeingsuggestedaS
a treatment for removal of arsenic and radioactive contaminants.
These safety precautions built into the McFarland plant can be
7
-------
useful to demonstrate and promote safe operation of ion exchange
treatment of other contaminants.
Level of Technology
The technology used in the mechanical design and planning for the
plant relies heavily on that used in the water softening industry.
The ...einical process design is based on research in use of anion
exchange resins completed under previous EPA grants. (Reference 1)
Aithougn that researc’ indicated efficiency may be improved by
further research on nitrate selective resins it was determined by
the principal investigator and the EPA Project Officer to use
conventional commercially available strong base anion exchange
resin as a basis for design. The plant would thus pro;ide base line
data against which improved resins and other process improvements
could be measured. The process did incorporate the partial
regeneration mode of operation whi had been m0nst t u erthe
research progranmentioned above and which is not used in the water
softening industry.
Loca t ion
Plant lcation at one of the well sites was dictated by the already—
in—place well and distribution syEtem which had been in use for
several years (over 30) and is typical for small communities
dependent on groundwater. Such systems develop over a period of
time usually by placing wells closest to the distribution system
5 jflg adeveloping area without regard for centta1 treatment.
McFarland can draw water Eros any of the six wells dhiCh can pump
water toan interconnecteddistributL0flSY5temamdt 5 5u9P t
to any part. Due to the lack of a centralized distribution system
the plant had to be desig’ ed to operate from a single well. Well
pumps operate on a demand basis, consequently, the niant i-ad to be
able to operate in an automatic on—off basis. This obviated the
need for on—site storage of water which would have added
considerably to the cost. The design was made to accept water
directly from the well pump, treat for nitrate removal, and allow
treatedwater
directing the treated water to a centraL part of the system and
without storage.
Capac ty
The delivery capacity of Well No. 2 is approximatelY 695 gallons per
minute (gpin) (lmgdonaCOflttnuOum basis). With nitrate levels in
the 15.8mg N03—N/L or 70.0mg N03/L range. A 6.8 mg N03—N/L or 30.0
mg N03/r.. product can be achieved by reducing nitrate to 0 in 70
percent of the water and blending with untreated water. Studies
(Reference 1) show a decrease regeneration ef tc1ency as nitrate
leakage in treated water approaches zero. See Figure 1. Since
8
-------
chemical costs are significant, savings can be raalizedbyuseof a
larger plant capacity and partial regeneration of the bed.
Furthermore, no reliance can be placed on nitrate levels remaining
inthel5.8 ngNO3—N/L3r70.0mgNo3/Lrange in the future. U these
v-sluesdid rise, treatment of allwater produced woulddefinitely be
required.
The plant was sized to treat the total well production rate and
provideablendingfacilitytoallowarangeof tr ’ atrnent levels from
partial to complete arid provide sufficient capacity to meet rising
nitrate levels.
Regenera tion Frequency
Anion exchange resins require regeneration ‘ith a sodium chloride
brine. It is necessary tohaveastnndbyregeneratedbedof resin in
a second vessel starting into operation when the first starts its
reger.eration cycle if uninterrupted service is desired.
Regeneration times (in McFarland about 120 minutes) are fixed
regardlessof bed s zewhile bed exhaustion t.imesor service periods
vary with size of bed and bed capacity (See Table 2). Bed exhaustion
timeshouldbelongerthanthereqenerationtjme if twobedsareused.
In the McFarland Plant with thzee beds in a staggered flow
arrangement this regeneration must be completed while one—half of a
bed is exhausted, i.e. the service period for one vessel must be over
240 ‘ninutes. The service period is a function of bed size, resin
capacity, and feed water concentration of nitrate, sulfate,
bicarbonate and chloride. In McFarland ’s plant variability In
water quality and capability to adjust bed depth from 3 to 5 feet
provides service time per vessel of 6 to 20 hours prov id ing adequate
time for regeneration.
Other considerations can also enter into de.erminlng regeneration
frequency. If an nsuEficient standby supply is available, it may
be desirable to increase bed size to allow for repair and ma intenance
of standby beds (repair of ma)or equipment inMcFarland requires at
least three working days). Or if only manual regeneration is
required it would be desirable to size beds for once per day erv ce
periods for operatc’r convenience.
In McFarland it would be prohibitively costly to size the beds for a
three day service period. If extensive repair and maintenance
cannotbecompletedwithin thestandbyperiod the systemcan be shut
down and other wells placed in operation or the plant can be operated
at at least 50 percent capacity if two vessels are operable.
Long standby periods require larger beds and added resin inventory
and equipment costs. Bed size iS also limited in that deeper beds
give higher backprcssurewhile large area bedsgive lower flow rates
9
-------
100
90
z
0
I—
, 80
D
4
I
60
4O
30 \
E
C)
Z 20
w
>
4
l0
‘ - I
0 5 0 15 20 25 30
L3S OF NoCI PCI CUBIC COOT RESiN
FIGURE I TYPICAL NITRATE LEAKAGE DEPENDENCE
ON SALT DOSE
10
-------
-- - -
TABLE 2 EST1: TED ::3. OF R [ CE ERATiO S PER VE L PER DAY
BeG VolumeS _____ Pr rcent Treated . a ter in Blend and Bed Cetn
Treated 1O 75 50’
3ft5ft 3ft5ft 3ftbtt 3ft ft
200 2.67 1.50 2.00 1.20 1.34 0.81 0.67 .40
300 2.00 1.20 1.50 0.30 1.00 0.60 0.50 .30
400 133* O.B0f 1.00 0.6 0.67 0.40 0.33 .20
* Tuelve hour service period per vessel and 6-hr stardby per vCsSCl
+ T :enty hour service period per vessel and 1O- r standby per vessel.
11
-------
(hydraulic loading) .ihich promote reverse adsorption or “dumping”
of nitrate from resin to product water.
Water Quality
Water quality is am extremely irnportant factor in nitrate ion
exhange technology. Two areas of concern are;
1. All ma or anions interfere to reduce bed capacity and change
product water quality.
2. Resin—ion equilibria and flow rate effects must be taken into
consideration to obtain proper bed operation.
Bed capacity, BV(N) or bedvolurnes treated per bed volumeof resin to
nitrate breakthrough, is s function of the concentration of all
anions in the feed water. Sulfate ion i the greatest offender.
Far the composition of eil 2 water given in Table 1, B’J(N) is
est imated as 351 bed volumes, if suif ate r ises to 200 mg/i and nitrate
to 16.9 rnq N03—N/t . ar 75.0 eq N03/L, 13V(N) is reduced to 228.
Chloridesustbe imcreasedoier 550eg/Ltohave the same ’ffectand
bicarbonateevengreater. Theinterfe renceofsulfateisthemain
reason for not building a nitrate demonstration facility at Well 3.
productwaterqualityischangeddue toremovalof influent ionsand
replacement by chloride. t ll sulfate is converted to chloride
throughout the run. Bicarbonate varies being low in the beginning
and rising toward the end of the run. This gives a product of
increasing alkalinity which is most corrosive at the start of the
run. See Figure 2. In the McFarland design water of low
cor rosiv ity is blended 1—1 with water of h icier corrosivity by us ing
two vessels in a staggered flow arrangement during service.
Another water quality consideration is the amount of chloride
increase inproductwaterduetoremovalofSUlfateafldflitrate. At
Well 2 the removal of 98 mg/i sulfate and 14.0 mg N03—N/L or 62.0 mg
N03/L. increases chloride by 106.5 mg/i; bringing the total to
approximately 200 mg/I.
Operational Sequence
The operational sequence is selected either through aprograannable
controller or manual push-button operation on each vessel. Each
vessel undergoes the following sequence:
1. Service. Raw water passes through the resin oed to exchange
nitrate and chloride ions until bed capacity is used.
2. Brine Injection. Asodiumchloridebrine ispassedthroughthe
bed to remove nitrate from the resin.
12
-------
C I
1 1C0 3
— PPU IU FC D
CUOLITE OI-D RESIU
ZELL 7W 3
. _ so 4
\:
-.- -.- -
— — -0--
N0
300
250- ____
200
U i
U-
—
0
100
0
0’
E
50
0
0 50 100 150 200 250 300 350 400
BED VOLUMES
S 0 B V
/
I
0-
I
FIGURE 2
TYPICAL BREAK THROUGH CURVE
-------
3. Brine Displacement. System water enters the bed to displace
br me.
4. Slow Rinse. System water is used to rinse last traces of brine
from rasmn.
5. Bac cwash/Resmn Declassification. Fines are removed from top
of bed and resin is mixed.
Under automatic contt ol any combination of valve operation can be
selected in one or more programs.
The resin declassification step is accomplimhed by a procedure
similar to the backwash with bifurcated inlets.
Brine Disposal
An important consideration i locating the plant at Well Site 2
rather than Site 3 was the lower quantityof waste brine generated.
Brine is disposed to the municipal waste treatment facility.
Review by the City and the State Water Quality Control Board was done
through the environmental impact report process. Because of the
quantity of dissolved solids added to waste water concern was
expressed for impact on soil and groundwater in the disposal area.
Treated waste water is discharged to 128 acres of agricultural lund
used for growing animal feed grains and cotton.
Thequantityof6%wastebrinedmschargedfromwell2wasestmmatedto
be about 0.5% of the ion exchange plant production or 5,Gi 0 gallons
per million gallons treated. If operated on a contmnuoui. 24 hour
day without blending this would require 2500 lbs. of salt per day
(raising TDS in the discharge area by about 600 mg/L) . Had Well 1o. 3
been chosen for the demonstration plant 2 to3 times this quantity of
brine waste would be discharged per day.
A monitoring p:o9ram was planned in the discharge area to monitor
soil, groundwater and wsstewater.
Automatic Operation
Automatic operation of the plant was considered essential to reduce
the amount of manpower required to sustain continuous operation of
the plant. Consequently, compleLe automatic operation was
planned. This automation was extended to automatic monmtorir g and
datacollection. Theonlymanuaj operations required were turning
the mystem on or off and inspection of datd and plant operation.
14
-------
Monitorina of Performance
Becauae this was planned as a demonstration plant and operating
costs and plant reliability were to be closely monitored ext nsive
plant monitoring was incorporated into the de .ign. The methods of
flow and bat:b.measurerient presented no difticulty, however, the
methodofco itintious nitrate monito 1ngw3 indoubtbecause of the
lack of reliable methods. Ion chromatography was chosen as the
method for test ing S inc it aoocst-ed adaptable to continuousni trate
monitoring and hdd the capability tomonitor other lonnof interest.
This is also an approved EPA nitrate method.
Planning for monitoring all flows and cycle changes was made. The
monitoring and control were to be performed by a progra.iimable
controller which could also give alarms and printed reports of
operation. Such extensive monitoring is not considered to be
necessary for a standard plant.
Flow Dt pm
A flo i diagram is shown in Figure 3. Feed water i3 supplied directly
from the well pump int two of the vetisels in the service cycle.
Vesseilbeing 50 percent exhausted when vessel2 starts its service
per iod -
When vessels 1 and 2 are in service No. 3 is in regeneration or
standby. After No. 1 is e hauoted 2 and 3 are in service, etc.
Service is stopped in any one vem sel by an electrical signal from a
flow totalizer or bj a manual signal.
The Brine Tank receives its water supply from system water supply.
Backflow preventers are provided. Rinse water and backwash water
areprovidedfromtreatedwater. Twoproportionatingvalvesserve
to blend treated and untreatea water.
Waste brine in any vessel is isolated from the main supply by Thiock—
and—blee& valves and check valves.
Electrical conductivity is monitored at well supply and product
water locations to detect any brine leakage into supply. Alarm and
shut down occur if product conductivity rises above that of supply
water. Nitrate levels are alsomonitored ir the blended supply and
can cause automatic alarm and shut down.
DESIGN AND OPERATING ?A AMETERS
rable3lis ts the bed characteristicsaiidthe target flowvaluesused
in vessel and piping design. The beds were sized to allow ample
service time for regeneration of the standby bed. The service time
15
-------
FIGURE 3 FLOW DIAGRAM
IO U EXCFfi UGE SYSTEM
-------
was estimated from the water quality (Reference 1) of the untreated
water assuming a partially regenerated bed thoroughly declassified
after regeneration andanitrate leakage of 30mg/i. Vesseiheight
wasdesignatedtoallowlO0%expansionof themaximumbeddepthtobe
used to allow for backwash.
during service.
Another water quality consideration is the amount of chloride
increase inproductwater due to removal of sulfate andnitrate. At
Well 2 the removal of 98 mg/I sulfate and 62 mg/i nitrate increases
chloride by 106.5 mg/i; bringing the total to approximately 200
mg/i.
Operational Sequence
The operational sequence is selected either through a programmaSle
controller or manual push—button operation on each vessel. Each
vessel undergoes the following sequence:
1. Service. Raw water passes through the resin bed to exchange
nitrate and chloride ions until bed capacit/ is used.
2. Brine Injection. Asod umchioridebrine ispassedthroughthe
bed to remove nitrate fro a the resin.
3. Brine Displacement. System water enters the bed todi.piace
brine.
4. Slow Rinse. System water is used to rinse last tracesof brine
from resin.
5. Backwash/Resin Declassification. Fines are removed from top
of bed and resin is mixed.
Under automatic control any combination of valve operation can be
selected in one or more programs.
The resin declassification step is accomplished by a procedure
similar to the backwash with bifurcated inlets.
Brine Disposal
An important consideration in locating the plant at Well Site 2
rather than Site 3 was the lower quantity of waste brine generated.
Brine is disposed to the municipal waste treatment facility.
Review by the City and the State Water Quality Control Board was done
through the environmental impact report process. Because of the
quantity of dissolved solids added to waste water concern was
expressed for impact on soil and groundwater in the disposal area.
Treated waste water is discharged to 128 acres of agricultural land
17
-------
TA JLL 3
BED CU A.RACL’w’ LSTICS 1 :.D T PC;:T rt.C S
Bed Cross Sec. Area
Bed Depth (Settled)
fled Depth )Lxponded)
Bed Volume
ALJPrOX. SaL. Urine Ccii.
Saturated Brine Rate
Brine Rinse Duration
Diluted iririC Race
BV of 6% Brine at above
Parameters
S1o Rìiase Rate
Slow Rinse 0uratic ri
Backwash Rate
backwash Druation
DeclassLficatiotI Rate
Declassification Duratron
arid procedure
Volume of resin i ii bottom
header *
Volume of resin iii straight
height
= 27.7 sq. ft.
= 36 in, to 60 in.
= 12 in. to 120 n.
= 05 cu. ft., 635.3 gal’
= to 142 Cu. ft., 1050.7 gal.
= 2.6 lbs/gal.
12 gprn to 43 7pm
= 20 i i i
= 63.5 ç,p t
= 1.5
= 64 ‘J to 110 ‘jpm
30 to 60 mini.
5 gpo/sq. ft. 140 gp.n
= 10 10111.
= Same as bacU .ash
= To be deterrniric ‘y o ecator
= 21.35 Cu. ft.
= 57.3 cu. ft.
13
-------
used for growing animal feed grains and cotton.
The quantity of o% waste brine discharged from Well 2 was estimated to
be about 0.5% of the ion exc’lancje plant production or 5.000 gallons
per million gallons treated. Ifoperatedon a Continuous 24 hour
day without blending this would require 2500 13$. of salt per day
(raising TOS ir the discharge area by about 600 ppm) . Had Well No. 3
been chosen foc the demonstration plant 2 to 3 times this quantityof
brine waste would be discharged per day.
A monitoring program was planned in the discharge area to monitor
soil, groundwater and waste water.
Automatic Operation
Automaticoperation of the plant was con”tde ed c sen’ia1 to reduce
the amount of manpower required to sustain continuous operation of
the plant. Consequently, complete automatic operatiofl was
planned. This automation was extended to automatic monitoring and
data collection. TheonlymanuaLoperations required were turning
the system on or off and inspection of data and plant operation.
Monitoring of Performance
Because this was planned as a demonstration plant and operating
costsandplant reliability,jere tobecloselymonjtoredextensi..,e
plant monitoring was incorporated into the design. The methods of
flow and batch measurement presented no difficulty, however, the
methodof continuous nitratemonitoringwas indoubt because of the
lack of reliable methods. Ion chromatography was chosen as the
rnethodfor
monitoring and had thecapabi.1.ity tomonitor other ionsof interest.
This is also an approved EPA nitrate method.
Planni”a for monitoring all flowa and cycle changes was made. The
monitorin and control were to be performed by a programmable
controller which could also give alarms and printed reports of
operation. Such extensive monitoring is not considered to be
necessary for a standard plant.
Flow Diagram
A flow diagram is shown in ?igure 3. Feed water is supplied directly
from the well pump into two of the vessels in the service cycle.
Vessel 1 being 50 percent exhausted when Vessel 2 starts its service
period.
When Vessels 1 and 2 are in service No. 3 is in regeneration or
standby. After No. 1 is exh 0 sted 2 and 3 are in service, etc.
Service is stopped in any one . sd by an electrical signal from a
19
-------
flow totalizer or by a manual signal.
The Britie Tank receives its water supply frum system water supply.
Sackflow preventers are provided. Rinse 4aterandbackwaShWater
are provided from treated water. Two proportionating valves serve
to blend traated nd untreated wo er.
Waste brine in any vessel is isolated from the maifl supply by “block—
and—bleed” valves and check valves.
Electrical conductivity is monitored at well supply and product
water locations to detect any brine leakage Lntosupply. Alarmand
shut down occur if product conductivity rises above that of supply
water. Nitrate levels are also monitored in the blended supplyand
can cause automatic alarm and shut down.
DESIGN AND OPERATING PARAMETERS
Table 3 lists the bedcharacteristicsand the target flow value ;used
in vessel and piping design. The beds were sized to allow ample
service time tot regeneiattonofthestandbybed. rheservicetime
was estimated from the water quality (Refercnca 1) of the untreated
water assuming a partially regenerated bed thoroughlydeclassified
after regenerattonanda nitrate leakageof 3Omg/l. Vesseiheight
was designated to allow 100% expansion of t” e maximum bed depth to be
used to allow for backwash.
20
-------
DESIGN AND OPERATING PARAMETERS
Table 3 lists the bed characteristics and tne target flow
values used in vessel and piping design. The beds were sized
to allow ample ser ice time foL rc’gencration of the standoy
bed. The ser:’ce time was estimated from the water quality
(Reference 1) of the untreated water assuming a part a1ly
regenerated bed thorouonly declassified after regeneration and
a nitrate leakage of 30 ity/l. Vessel height eas designated to
allow 100% e’.pansion o the maximum bed depth to be usea tp
allow for beckw sh.
HEADWORKS
The lower left portion E Figure 3 shows flow and piping for
the headworks. A general view of the plant is Shown in Figure
4. The well pump (7a lip) is in the center foreground. Thc ion
exchange vessels and tue laboratory building art? shown in the
backyrouno. Tne main line from the plant is shown in the lower
right corner as it enters the underground distribution system.
Tne open pipe on the tar right is an open discrarge for
flushing the well after long periods of inoperation. The
valving of the neadworks is not visiole in this view.
The pump discnarge line is shown in the foreground of rigure 5.
Water leaving the pump enters the system turougn the check
valve and thc butterfly valve before entering the booster pump
shown on the left in Figure 5. A butterfly valve to puss water
directly to the distribution system is also shown in this view.
The 8—inch main supply line is shown in Figure 6 running from
the plant to the underground distribution sysrem. In this line
a main shut-off valve, a water meter, two rnercoid pressure
switches and a shut—off valve to the plant are shown. The
mercoid switches shut the plant down when di.strtoution system
pressure is high (over 65 psig). The second switch is a backup
for the first. Power for the pump is restored five minutes
after the pressure switches arc activated by low pressure (55
psig). Also the booster pump is shown in this view in the
upper right. Details between the booster pump and the ion
exchange vessels are shown in Figure 7. Figure 7 shows the
booster pump on the right with the steel discharge line Joining
the plastic 8—inch line with a flexiole coupling. Thus line
makes three right angle turns before entering the feed manifold
to the vessels. A nercoid pressure switch and pressure gauge
are connected into the plastic line. The mercoid pressure
switch is activated at 100 psig to shut power to the pump off
in case flow through the vessels produces a back pressure of
iuO psig or more. Power is rebtored in this case by a push
21
-------
I ’.
•,
FIGURE Li. GENERAL VIEW OF PLANT
22
-------
-
r
FIGURE 5. PUMP DISCHARGE PIPING
23
-------
button switch on the power panel. In this view the distribution
line from the plant is shown in the center of the picture
crossing from left to right. A motor driven eutterfly valve is
shown in the center of Figure 7. This valve can b/—pass water
from the booster pump directly to the distribution line and can
be automatically operated on signals from the microprocessor.
These signals in turn are controlled by nitrate levels of the
well and product water from the ion exchange vessels. A wash
water line is seen co r ning from the distribution maii and
running to the wash water manifold to the left of the t’ lcture.
Figure is a clobe—up view of the motor operated by-pass
valve. The distribution line is in tne foreground, the booster
pump discharge line is in the background. The ion eAchar.ge
vessels are out of view to the right ot the picture and the
pumps are out of view to the left. The backwash and rinse
water line is shown at the top.
Figure 9 shows most of the above features in a general VtL4.
The motor operated valve is in the center of the picture. The
pump discharge line can be seen running from the booster dump
to the vassels on the right through the line with three right
angle turns. Tne main oistribution line runs from the vestels
on the right without any aends to the left. On the rignt side
of the picture a meter box can be seen which houses flow meters
for wash water, treated water, and backwash water. The brine
tank can be seen in the left bdckyround.
ION CXCUA GE VESSELS
As shown in the flow diagram (Figure 3) three ion exchange
vessels of equal design contain the resin beds and tne
appropriate valving. The vessels have a straight side height
of ten feet to accomodate resin beds of up to five feet deep to
allow for 100 percert expansion on backflush. Figure 10 is a
general view of the vessels showing the feed water, product
water, and wash water manifolds on the far left, the vessels
with a control panel on each, protective canopy and concrete
pads. The vessels are placed on a concrete pad lO’x30’x12”.
The utility pad between the vesse 1 s and the laboratory building
is 30’x30’x2”.
Fiqure 11 is a close—up view of the upstream side of vessel No.
1.. Two pipes are seen on the left side of the vessel branching
from the manifolds. The upper 4—inch pipe is the feed water
sufply and contains a batch meter. The lower pipe is the
backwash water supply. The “Z shaped pipe parallel with the
left front leg is the product exit pipe from the vessel.
24
-------
FIGURE 6. PIPiNG TO DISTRIBUTION SYSTEM
25
-------
FIGURE 7. DISCHARGE FROM BOOSTER PUMP
26
-------
FIGURE 8. MOTOR OPERATED BLENDING VALVE
F”
• ,._i
S., -
27
-------
FIGURE 9. GENERAL VIEW OF HEADWORKS PIPING
28
-------
FIGURE 10. FRONT OF ION EXCHANGE VESSELS
29
-------
Figure 12 is a view of the right hand side of vessel No. 3.
The control panel is on the left in front of the feed water
pipe and the backwash valve and pipe. On the right of the
vessel various pipes run from the manifolds in the back to the
front of the vessel. These pipes from top down are: a 1—inch
brine line and automatic valve, a 1—inch slow rinse line, a 3—
inch backwash exit line. The lowest line is the 1—inch product
line containing two automatically operated valves. Tne waste
discnarge lines are shown in the lower right over the grating
of the waste collector oox.
Figure 13 is a view of the rear of the vessels with the pumps
and headworks on the right. Th ’ various manifolds connected to
each vesz.el are from the top down are:
1. Brine inlet 1”
2. Slow rinse inlet 1 ”
3. Wash water and brine exit 30
4. Feed water inlet 6 ”
s. Wash water inlet 3 ”
6. Product water line 6”
VESSEL INTERNALS
The internal piping of the ion exchange vessels is representeo
by the drawing in Figure 14. The internal elements can be
identified by the outlet number. The drawing also gives the
dimenQmons measured from the top of the base of the center hub
of the bottom drain. No. 1 is the brine distrioutor. As shown
it is positioned to deflver brine six inches above a three foot
deep resin bed. Rotation of the distributor 180 degrees aoout
the inlet axis positions the distributor for a five foot deep
bed. Nos. 2 and 4 are the drain outlets which drain product
water from each half of the spoked collector, No. 5. Element
No. 3 is the feed water distrioutor. Not shown on this drawing
are the various openings in the side of the vessel to
accomodate satplers and other devices which may be desired in
the future.
BRIdE SYSTEM
Piping for the brine system is shown in the upper left of the
flow diagram (Figure 3) and in Figure 15. Water is added to
the brine tank directly from the systen via the 1.5—inch pipe
shown at the left in the picture emerging from underground
running to near the center and upward into the brine tank inlet
(not shown). Flow into the tank is controlled by an
electrically operated valve and can be activated from level
30
-------
FIGURE ii. VESSEL PIPING - LEFT SIDE
31
-------
FIGURE 12. VESSEL PIPING - RIGHT SIDE
32
-------
FIGURE 13. REAR VIEW OF ION EXCHANGE VESSELS
/
*4 -
‘ ‘ •••.
33
-------
NOTE: DIME lSIONS FROM TOP
OF BASE PLATE
FIGURE 14 INTERNALS OF ION EXCHANGE VESSELS
34
-------
sensors or can be operated manually from the main control
panel. Liquid level i.n the tank can easily be observed with a
2—inch d ameter sight glass. Tne latter means of control was
used during this period of operation since an operator made
frequent visits to the site.
The outlet of the brine tank can be seen in the upper portion
of the picture. Concentrated brine is drawn through this 2—
inch pipe by the small centrifugal pump shown. Brine is pumped
through the 1—inch piping on the right through a check valve
and an electrically operated solenoid valve. The latter was
added after fa lure of the check valve to close on several
occasions. The electric valve is powered oy the same circ.uit
supplying power to the brine pump.
Concentrated brine is blended with water by the piping shown in
the foregroind before it enters the underground pipe feeding
the Urine systems of the vessels. Diluting the brine was felt
to favor the dischage at waste brine to the local sewer system
and to prevent resin deterorizution oy osmotic shock.
PRODUCT DELIVERY . YSTE 4
Tne product delivery system is shown in the above figures
showing the vessel manifolding arid headworks. Product is
collected from each of the operating vessels arid allowed to
flow directly into the distribution main. The product can be
blended with untreated watCC using the blending valves
provided. Blendiricj can be controlled by automatic means
through the microprocessor or manually at the discretion of the
operator
PROCESS CONTROL
t4anual Control
Due to the research nature of this program i.e was decided to
provide for manual operation of the plant to work out an
algorithm for the microprocessor program. This was especially
desireable since the process employed was new. Furthermore,
resin declassification had not been previously employed in a
system of this type and valving sequences had La be laboriously
determined by a trial and error method. Furthermore, this
allowed the plant to be placed into operation before all
automatic devices including the microprocessor arid main control
panel had been completed. The wisdom of this choice became
more obvious when it was found what trouble and expense were
involved in programming and testing the microprocessor control
35
-------
FIGURE 15. BRINE SYSTEM PIPING
-------
system. The approach here wan to work out in as much detail as
possiole the program by trial and error testing under manual
control; then write the computer program acc.rrdinyly.
Manual operation was accomplished by controlling the valves on
each vessel from the push button control panels. Flow
quantities were measured by several meters and flow rates were
adjustable by hand valves except for service flow rate which
was only adjustable by blending.
Service flow was metered by batch reset meters on each vessel.
Figure 16 is a view of the batch reset meter in the feed water
line to vessel No. 1. A pulsed signal is sent to the box above
the meter which advances the register in the small box shown
directly above the meter. When the registered gallonage equals
a preset quantity an end—of—batch signal is sent to the control
panel. A red light indicates to the op—rater that the vessel
needs regeneration. A wiring option inside the control panel
also allows the s gnal to take the vessel out of operation.
The regeneration cycle for sne vessel can be completely
controlled frost tne panel. Figure 17 shows the meters which
measure the amour’t of concentrated brine, dilute brine and
make—up water for brine. Figure l B shows meters for total
treated water, backwash water, and rinse water. After brine,
rinse, and backwash ore completed the vessel cart be placed in
stand—by or into service.
SEMI-AUTOMATIC OPERATION
After flow rates are adjusted the regeneration off each vessel
may be performed by an automatic timer/sequencer which was
installed in each control panel. These timers were of the type
used for controlling irrigation systems (Rain Bird Model CRC—
B) . The operator was required to initiate the regeneration
cycle and place the vessel into operation when required. This
required the operator to make frequent checks or estimate the
end of service for the next vessel.
AUTORATIC OPERATION
Automatic process control is accomplished witn a TI—500
microprocessor manufactured y Texas Instruments and the inputs
from the flow meters, conductivity sensors, and the nitrate
analyzer. Figura 19 shows the main control panel whxch houses
the process controller and into which all of the control
signals are sent and some of which are displayed. Nitrate
levels are recorded by the top recorder. Electrical
conductivity of well and product water are both recorded on the
bottom chart. The light b04 annunciator on the top right
37
-------
FIGURE 16. BATCH RESET METERS
37a
-------
FIGURE 17. BRINE AND MAKEUP FLOW METERS
38
-------
FIGURE 18. TREATED, BACKWASH, AND RINSE
WATER FLOW METERS
39
-------
displays the status of each vessel and pumps. Conductivity of
waste water, well water, and treated water are displayed in the
three read-outs below the annunciator. Two read—outs above the
switches give total water del.ivered and treated. The switches
turn on the microprocessor, select the vessel sequence, control
water to the brine tank, and reset after alarm. The printer
gives a daily report of all flows, nitrate levels,
conductivities, status of each vessel, and alarm conditions.
Reports are also printed at any time on demand and are
automatically given for alarm conditions: high nitrate, high
conductivity in product and power failure.
Figure 20 shows the interior of the control panel. The
microprocessor equenceL and power supply re the two white
boxes on the right side. The input modules are on the lefL.
Wire terminals are on the bottom. The small box near the
sequencer is for inputting numbers to the ‘nicroprocessor such
as brine dosage and service oatch used in the control program
or in the calculations for the report.
Figure 21 shows the ion chromatograph used for nitrate
monitoring. The unit shown is a dual unit. One is dedicsted
to nitrate measurement of treated or blended water. The other
unit is used for research purposes. A signal of nitrate level
goes directly to the control panel. The upper left compartment
is the chromatography unit. The pump module5 are on the lower
bottom. The upper right houses the conductLv.ity meters; the
automatic controller is below these.
Figure 22 shows the interior of the 12’x24’ ouilding being used
as both laboratory and control room for the plant. The ion
chromatograph is shown center right. The enclosure benind the
instrument contains a store of distilled water and other liquid
solutions for the operation of the analyzer.
Figure 23 shows the interior of the other end of the building
with the above described control panel to the right of the
door.
40
-------
FIGURE 19. MAIN CONTROL PANEL
tffl
V 4
41
-------
FIGURE 20. MAIN CONTROL PANEL INTERIOR
42
-------
FIGURE 21. ION CHROMATOGRAPH
43
-------
0
k
FIGURE 22. LABORATORY AND CONTROL BUILDING
INTERIOR WEST SIDE
44
-------
4 _____
FIGURE 23. LABORATORY AND CONTROL BUILDING
INTERIOR EAST SIDE
45
-------
SECTION 4
PLANT OPERATION
CHECK OUT AND START UP
Resin Purchases and Loading
Resin was purchased witn funds prw’ided oy the CPA, Kern County
Water Agency and McFarland Mutut i Water Company. The resin
purchased was Duolite A—bID and was sufficient for the initial
three foot bed depth (225 cubic feet) plus the additioral two
feet of bed depth (170 cubic feet ) and 50 cubic feet as make up
resin. All resin was purchased at one time to avoid future
price increases. See Table 4.
TABLE 4
RESIN PURCHASES
Order No. Quantl.tI Price Agency
1 18 cu. ‘-• $ 2,371.68 McMWC 0.
1 237 cu. ft. 31,227.12 FCWA b McMWC0.
2 170 cu. ft. 22,399.20 EPA
2 50 cu. ft. 6,588.00 EPA
The purchase price of the resin was $124.30 per cu. ft. plus
$7.46 sales tax. Prices include delivery to the job site.
The resin was ordered and delivered in drums each containing
five cubic feet. Handling drums for unloading or loading into
vessels requires five men, one of which operated a fork lift.
To load the vessels, a temporary wooden frame was constructed on
top of the vessels being loaded to allow the drum to be rolled
over the top manhole. Two men positioned the drum onto the fork
lift which was operated to place the drum near the frame. Two
men on the cat walk placed the drum on the frame and rolled it
over the open manhole and opened the drum and plastic oag to let
the resin fall into the vessel. The vessel was partially filled
with water to prevent damaje to the resin as it fell into place.
46
-------
This operation required one hour to load fifteen drums (85 Cu.
ft.) into each vessel.
Salt Delivery
Salt was delivered to the site approximately one week after
placing the order with a local distributor. The delivered price
was $31.50 per ton. A load of 25 tons was transported from the
San Francisco Bay area. Salt unloading required approximate’y
one flour. The only labor required was provided by t ie driver
and his cost was included in the delivery cost.
Plant Check Out
Prior to resin installation the ion exchange vessels and
plumoirig were pressurized to check plumbing leaks and valve
operation. Testing continued aJ er resin was installed. The
following malfunctions were noted:
1. Automatic valve closures were found to oe too rapid in some
cases and caused water hammer pressure rises within the
system. These were adjusted.
2. The lntergrating flow meter on Unit 1 could not be reset.
This was replaced.
3. Some flow meter sensors were malfunctioning. Readings of
treated water flow rates were off by a factor of .7
according to other flow meters in the system. This flow
sensor was replaced.
4. Plumbing leaacs were noted in the manifold lines and brine
tank plumbing. None were serious.
5. It was noted that when all three vessels were in service at
low flow rates (approximately 100 gpm each) the flow
channeled through only two of the vessels and stopped in the
third. This is not a malfunction and can be expected at low
rates. No operational mode will require this type of
operation.
6. Pump discharge pressure appeared to be nigher than expected.
This was thought to be due to a rise in ground water level
over the previous year. This has caused high pressure build
up on the inlet side of the vessels and breaking and severe
leaking from the quick connect plastic caps (3—inch) on the
top of each vessel. These caps were replaced with aluainum
caps. They were eventually replaced with blind flanges.
Pressure drop across the ion exchange vessels was
47
-------
approximately 10 psi..
7. On backwash with resin in tal1ed the filter in the backwash
discharge line was immediately c1og ed with resin fines.
This fslter was removed and the backwash was allowed to fLow
through a 50 mesh screen and into the sump to catch the
fines. It was noted tnit approximately 300 ml of fines were
recovered on each of the initial backwashes. These fines
were eventually washed from the system and were reduced to
zero.
8. The pressure limiting switches were not fuictioning to allow
automatic well pump control at low and high distribution
system pressure. These switches were faulty and
recommendations were made tc replace them with Mercoid
switches which were found to be very reliable. Manual pump
on—off control was reuqired until these limiting switches
were replaced.
Resin Washinq
After the resin tanks were loaded and sealed each bed was washed
pith water availamle from the distribution system. The resin
manufacturer recommended either washing or chessical treatment to
exchange all ions in the resin. The washing method was
preferred to save chemical costs and disposal problems and
because of the availablil.tty of a plentiful water supply. If
sulfuric acid were used for this operation, approximately 75
gallons of concentrated acid and 1300 pounds of sodium chloride
would be required for one complete ion exchange. The number of
ion exchanges required was not known. It was decided to try the
washing method first.
The washing method was accomplished by leaving the well pump off
and placing each unit in slow rinse. In this manner each bed
could me rinsed at a rate of approximately 50 gpm (0.08 13V per
mitt). fowever, it was found that only two vessels could be
rinsed simultaneously in this manner. Slow rinsing was
recommended by the manufacturer since both rinsing time and
quantity of water are required to remove the excess trimethy].
amine from the interior of the resin beads. Approximately
300,000 gals (470 8V) of water were required to eliminate the
taste and odor of trimethyl amine in the rinse water for each
bed. This operation required approximately 100 hours for two
vessels and an additional 100 hours for the third vessel.
The water used for rinsing was the water which happened to be in
the distribution s 4 stetn . Wells No. 6 and I were in use at the
time. A chemical anayi.sis of the entering and leaving waters
were obtained. See Table 5. The total anion content of the
48
-------
wash water was 2.7 meg/i. Consequently it appears that
approximately one ionic equivalent is required par volume of resin
for the resin conditioning. (1300 ‘neg/l divided by 2.7 meg/i 481
By) r s pointed out above, 470 BY were used.
TABLE 5
WASH WATER COMPOSITION
Constituelt Milligrams Per Liter
In Out
Carbonate 13.6 0
Bicarbonate 38.1 36.6
Chloride 23.0 74.4
Sulfate 50 5
Nitrate 2.8 rig N03-N/L Less than .1 N03-N/L
or 12.4 mg NO3JL or .4 tug N03/L
Arsenic 0.04 Less or . 0.01
pH 9.2 7.8
The arsenic in wash water was from Well No. 6 ’rhe nitrate was from
Well No. 1. It is interesting that the resin was very effective in
removing the arsenic as well as the nitrate and ha. the added advantage
of reductn4g the pH to a desirable level. Well No. 6 arsenic levels
have been found to be as high as 120 micrograms per liter and the pH is
well over 9.
The above analysis was taken at a By throughput Df 405 By. If arsenic
is being removed as a divalent anion, probably HAs 0 = at pH9 it is
expected to treat Well No. 6 water and obtain over 800 By of arsenic
free water per service cycle.
Initial Regeneration anJ Washing
Before automatic operation was initiated it was necessary to test
performanceusing theoperational parameters, someof which aregiven
in Table 3. Tiese include flow rates, pressures, duration of wash
and backwash, method of resin declassification, breakthrough
characteristics, and water quality parameters. These studies
could best be done while operating one vessel at a time. This also
allowed the establishment of methods of manuaL operation which may be
necessary or useful at times in the future. Each vessel was prepared
for one—at—a- time operat ion by regenerat ing each bed with 1.5 BV of 6%
brine (or equivalent) to remove sulfate, arsenic, and sulfate
accumulated during the initial washing.
49
-------
Sample ports are located at the same positions in the center of
each bed. The location of each port in relation to the upper
resin surface is given in Table 6.
TAdLC 6
POSITION OF SAMPLE PORTS
Port No. Inches Below Resin Surface % Into Bed
6 (15 inches above surface)
S ( 6 inches above surface)
4 1. 2.72
3 1 19.01
2 14 36.02
1 28 76.04
0* 36 100.00
* 2ort 0 is same as vessel outlet.
The initial regenerations were accomplished with some deviation
from the above target values. It was fDund that diluted brine
rates varied from 22 to 36 gpm with the result of using a more
concentrated brine than expected. This was due to a flow
restriction valve in the waste brine exhaust. This was
repaired. The effect was to requiie more slow rinsing than
expected. Rinse duration was 60 minutes and used up to 3990
gallons (6.3 1W) . Table 7 shows the drop in TDS of the wash
water throughout.
TAdLE 7
TDS OF WASH WATER
Sample TOS Gallons BV
Fresh
brine 150,000 0 0
15 m m 4,000 960 1.51
30 m m 800 2020 3.17
45 mm 600 3000 4.72
60 twin 474 3990 6.27
Lesser wash water quantities were used as fresh brine
concentration was reduced to 6%.
50
-------
flu t i al Colurin Exhaus ti ons
The columns were exhausted one at a time with Well 2 water. The
main by—pass valve (rather thai’. the nlnding valve) ias uued to
by—pass part of the pumped water. The plant was started using
the following procedure:
1. Well pump OFF
2. Power to selected vessel turned ON
3. Control switch set to NhNU/ L
4. Service switch turned O
5. Main by—pass valve 3pened full to by—pass plant
6. Turn well pump ON
7. Partially close by—pass valve until desired flow rate
through vessel is achieved.
8. Readjust ny—pass valve until desired ratio of treated to
untreated water is achieved.
Tne plant was tnen allowed to run wh i le monitoring nitrate
levels in the various sample ports until the nitrate level
reached 30 ng/l in the effluent. At this point the following
shut down procedure was employed:
I. Well pump shut OFF
2. Power tO vessel turned OFF
3. Regeneration switch turned ON
4. Allow vessel to remain in stand—by until start of
regeneration.
NOTE: It is important to shut well punp off oefore turning
vessel OFF to avoid high pressure surges into the vessels.
The next vessel was placed into operation with the following
procedut e.
1. Power to vessel turned ON
2. control switcn set to MANUAL
3. Control switch set to SERVICC
51
-------
4. Well pump turned ON
5. Ad]ust by—pass valve if required to obtain des ired olend
A chemical analysis of untreated, treated and blended water was
obtained from a certified laboratory. The results are yiven in
Table 8.
TAOLE 8
CHEMICAL ANALYSES FOR INITIAL RUNS
Constituent (mg/i) Untreated Treated blend
Calcium 94 96 96
Magnesium 2.1 2.1 2.1
Sodium 68 68 68
Potassium 3.8 3.6 3.8
Bicarbonate 119 27.7 120
Chloride 88.2 264 166
Sulfate 120 less than 5 40
N03—N/L 16.0 2.6 6. 6
NO3/I 7L).9 11.5 ng 29.2
mg
Color Units 1 1
Odor none none none
pH 7.2 7.4 7.4
C, microinhos 790 920 850
DECLASSIFICATION TESTS AN0 MANUAL OPERATION
The ion exchange p1 mt was operated manually using one vessel at
a time at the approximate flow rates described above. Tests
were performed by reducing the amount of brine dosage for each
subsequent regenerat on until asignificant nitrate leakage was
obtained. Salt. dosage equivalent to one BV of 6% brine gave
significant nitrate Jeakages to test the resin declassification
system. Subsequent runs were then made at that dosage while
changing the declassification cycle parameters.
The method of declassification which appears to be most
effective is to backwash with only one of the two (V—106 and V—
107) backwash inlet valves open for a three minute period and
then closing for a three minute period to allow the bed to fall.
This sequence is then repeated a number of times to further mix
the resin. The effective.iess of the operation can be assessed
from the shape of the breakthrough curve taken during the
52
-------
12 NITRATE
SAMPLE NO. ( T EA.15.28V)
FIGURE 24. RUN NO. O1114-P3
100
50
0
z
0
z
4
U
-J
U
a.
0.... 5... .10.
53
-------
z
z
U i
I-
-J
U i
a
tOO
] :NITRATE
SAMPLE NO. (AT EA. 5.2 BV )
FIGURE 25. RUN NO. 12143—Pi
50
0
0... .5... .10... .15..
54
-------
Suoseque.it run.
The data for each run was entered and analyzed by a spread sheet
program. Graphs of the breakthrough curves were also obtained
from the sPread snect data. The breakthrough curves aie given
in bar chart form. These typical spreadsneets and charts are
given in Tables 9 and 10 and Figures 24 and 25. These represent
the best and worst cases respectively. The analysis of all test
runs s listed in Table 11.
The first three runs listcd in Table ii. were regenerated after
the initial resin washing with system water which contained
nitrate. Trie beds were not fully exhausted with sulfate or
nitrate. Although both halves of each bed were bac washed
separately, poor mixing resulted as seen in the corresponding
breakthrough curves. The experiments continued with a series of
regeneration/exhaustion cycles starting at high salt loading and
decreasing oy 0.5 By (6%) on subsequent runs. The reason being
to achieve a hign nxt.rate leakage where seni.itivity to
declassification is greatest. It was found the system could be
operated using only one 1.0 By (ba) regencrant. With no attempt
to mix resin very poor results ere achieved in run 12.L43—P1.
Suusequenr runs snowed gradual improvement. It is apparent from
the results that simple back4ashiny does not mix the resin
sufficiently even if only one bac . ..4ash valve is used.
That the intermirtant on—off cycling is effective indicates
resin is missed durx g the start (or stop) of the uneven
backwe sh period. At tne beginning the bed is not fully
fluidtied and resin agql.tation is quite oovious as viewed
through the sight jlass. At the end of the cycle the upper
portion (cnioriae enriched) of unfludized bed may redistribute
itself on the bottom of the bed before tne fludized portion of
the bed falls to compact tnus achieving an overall
redistribution of resin in the beds (and chloride enrichment at
the bottom of the bed)
ANALYSIS O INITIAL OPERATION
The data of the above runs provides the initial opportunity to
compare actual with estimated plant performance as far as salt
usage is concerned. All data from the breakthrough curveb were
entered on spread sheets and regeneration effic:encies ottained
from the calculations within the spread sheet. A summary of
important performance data is given in Table 12.
The estimates of nitrate leakage and BV( ’i) were made using the
method developed and reported under the previous grant.
(Reference i) In general nitrate leakages are considerably
55
-------
lower than the estimates and actual OV(N) values are 13 to 15
percent lower than estimated.
The latter may be due to stopping the run prematurely at 33 mg/l
and not taking into account the resin capacity expenditures
during washing. Approximately 4.5 av washing were used in the
slow rinse cycle. The lower leakages and shorter bed life
translates into a considerably higher salt efficiency than
expected fron the estimates. The nitrate removal efficiency is
about 75% higher than expected as indicated by the data in colum
7 of Table 12 for the runs using only 1.0 BV (6%) regenerant.
A further indication of salt efficiency and effectiveness of
partial regeneration is the ratio of the equivalents of chloride
used per equivalent of nitrate removed (Brine Use Factor). This
data is summarized in Table 13.
The higher salt efficiency obtained indicates that the
concentration of nitrate by chron atographic action on the column
should be taken into account when estimating salt removal
efficiency. The present modeL used in estimating this parameter
assumes an even iistrxbution of nitrate throughout the bed. The
estimates of BV(N), however, are made 4ith the chromatographic
effect taken into account. It is also likely that lower
leakages are obtained because the columns had not reached
equilibrium, i.e. they had not been operated repeatedly at the
same salt dosages before the data were obtained. This data
should be compared with that obtained under repeated and
continuous operation in Section 5.
56
-------
TABLE 9
DATA FOR RU 01114-P3
ru.i
:., i
dat -a i
H = 1114—P: ;
I -R 10 1. ]
LI 11 Ho=2
Lit r;i
F: ii
L.i t:r .
Er i
:::
Iii 9
: ij4 F
L i i 2 Tj
212.0
2.53
= 1.14
r - S+14= .3 .54.
P r- 1 itr ji lu vit.
rr — - c-.-
I I — •.i :.. ._i
1 .4
b :d .,cd t; tt - i i
P - .r
ULU.W
U. U
0. U
U.U
U. U
U. U
15.2
• 4
;c i . :
I ).
91.2
121 .
1 T 2
i:;2. 4
1:3; -. S
212.3
cal
i3
0
P .r litEr
Pi :- 1
pI. r- l-it ar•
r 1 - ;l
&.‘.:- .J P.at-
1’ -i j.
r -r.j i
4.
4.
24.5
• b
i9•. 4
13.4
4,4
0
i . 4
14. :1;
23. 4
22.3
24. ‘5
33-. 9
c pp -
i - , - t
I43::ct P
ut =113
C t
E r- ±
Fr .c—
t±Ô i t . .9J-tiaI
F: -1 T
-------
TABLE 10
DATA FOR RU 1 12143-Pi
— _-,1l , -
.Uu i 4 ..
= A lUlL;
IJ ill Flo= 2
Lit; r’
of
0-.-
i-. .z.J.ii — . .-ft
of 6
=24U1
•4 33 p Ii
i1 1 =
ili
C l PPrc
i . i . i = 32.2
HCC’2 Pr i
i
I-to of
Br-ii
Fr .c —
tIOfL p - -tj. l
F:iJH
FUF. liRT
UCjIJ. Iiii:’l. i 4U
P r lit. r
rc:- - 1 ii
Ii 4 C 1
P r
?
rr r’ ov . =
1 .:i.I r - 1
of
t•IJ.
ga t. .
ill
-. — .. 2. -.
vU L 1 1l 1 ...IJ.L ’_
.u.lf i.t ’ c . .l : ri
P . .u -J.r , tc -r-
‘.i. . . 1’. ’. .
A c._i •
uuu. Li
U • Ii
0. ci
U. U
U. U
U • U
U • U
15.2
30.4
45. £
:
C l
9-1 . 2
10’:. 4
1_i.
12i: . :3
145.13
1:::. 13
2
4 . . . I. c_I
.1 • .4
41.4
41.4
41 . 4
41.4
- 1 -4 -.4-
41.4
3 . 1
9.7
13.13
U. U
I- ,
• i _I
4.2
I I.
U.U
4
I - ,
i—i. ‘.2
13.2
21.
• . 2 . . - .
11 ’ 11 -q 131.
u rJ
Pc .’r 1. it. -r ’
r -’ - i° i•• i = ic;üc;.
p I: ’ ri
12L 3
= 113
=
rr, q :3134 = 2.
H132 = 1.14
i— ’. tt — .
i fit •
P r litr iiutlu -it,
1-4133 -
= 595.
L1:th)
P P ii : ii • 4 - . -
- .
L iL1’ r - L t
58
-------
T7\I3LE 11
EFFECTIvCNrSS or DECL1\SSIFIC? TION
_________ Nc2hod of Dcc a sificn1ion
Resin .:as not fully loaded on
fir L thrcu runs. )ac1:u ish done
for 5 mm. V—lOG and 5 mm. V—107.
Same backwash as above. Unsteady
lcakocjo.
Nc attempt to mix resin. No backuach
Backwash only for 10 mm. with no
nttcmpL to mix resin.
V—lOG only for 16 mm.
V—lOG only. On three mm. off
3 mm. Three times
01114—P3 V—lOG only. On 3 mm. off
3 rain. i our tirres.
Run No .
1 1013—Pi
11 023—I’2
1 1093—P3
11153—Pi
11 173—P2
U,
0
11223 —P3
R . sult 1W 6% Salt
12143—N
122 13—P2
Not 1.5
Satisfactory 1.5
1.5
Not 2.5
Satisfactory
Poor 2.0
Lo., initial leakage 1.5
lEigh second half
Very Poor
Improved 1.0
Better than 12213-P2 1.0
-------
TABLE 12
COMPARISON OF ACTUAL AND
Run No .
110 13—Pi
110 23—P2
1109 3—P3
11153 —P 1
11173—P2
1 1223—P3
1214 3—P1
12213—P 2
01 114—P3
DV (N)
C31c Pctuel
251 243
251 250
251 259
270 244
266 250
251 220
229 198
229 199
229 213
Le3kage
Nitretc MOcT/1
C ]c. AcLu 1
29
29 3
29 7
16
21
29
10
11
8
42 11
42 10
42 17
Ncq N Removed
Per liLer resin
Ca c. AcLti 1
170 241
170 274
170 270
244 242
215 241
170 231
107 190
107 195
107
2.5
2.0
1.5
1.0
1.0
CALCULATED PERFORMANCE
Selt
BV lb per
6% Cu. l’t
1.5
1.5
1.5
5.61
5.61
5.61
9.35
7.40
5.61
3.74
3.74
184 1.0 3.74
2” column 311
313 (VIRGIN RESIN)
-------
Th!3LC 13
REGENERANT Er’rIcIENcy — PLAUT DATA
Run No .
1 1 01 3— 1 ’ 1
110 23—P 2
110 93—P3
111 53—Pi
11173-22 8.30
11223—23 6.50
12143—21 5.26
12213—22 5.13
01114—23 9.35 5.45
= Neq Chloride Used in flr ne
* Brine Use rector
Ioc nirroLo rciiovcd from fccd Y LC:
Sci.1 t
RV of G NaCI
Brine Use Factor*
CdlculrItcd I\CLU . J .
6.82 6.22
1.5
1.5
1.5
2.5
2.0
1.5
1.0
1.0
1.0
5.47
5.56
10.33
8.82
8.62
10.25
9.30
0.02
9.35
9.35
-------
SEMI-AUTOMATIC OPERATION
A period of semi—automatic operation followed the above tests while
the TI—SOD Controller was being installed and programmed. During
this period irrigation type controllers were installed in each
controlboxtoautonatethereqenerationcycle. Thlswasdonemainly
for operator convenience. Manual control was required to initiate
the regeneration cycle and to control the brine pump. The ma n pump
wasaisounder manualcontrol. Nitrate levelsof theeffluent from
each vessel and blended water were also done manually during this
period while tne analytical chemistry instrumentation was being set
up and tested. This period was also used for operator training
although the operation was more complex than it eventually became and
more analyticalche’nistrywas requiredof the operator for monitoring
purposes. Also during this period only one vessel wasoperated at a
time to further simplify the tasks of the operator.
Data was collected for each vessel exhaustion with service period
terminated when nitrate leakage reached 30mg/i. These runs were
made to determine the bed capacity per service cycle.
AJTOMATIC OPERATION
Pre—Start Tip Checklist
Automatic operation of the plant was accomplished with the TI—SOD
Contiolier.
Prior to starting the plant in automatic operation the following
steps must be followed:
1. Determine the sequence of vessel operation Si, 2, 3 or 2, 3, 1, or
3, 1, 2) and set the switch on the controller accordingly. The
lead vessel (the first to be exhausted) is designated by the first
number. The vessel in standby is designated as the last number.
If the plant has previously been in operation, the lead vessel. is
determined as the one showing the highest flo. reading on the
batch register.
2. if the resin in each vessel is regenerated, the end—of—batch
should be set at one—half its value for the lead vessel and at the
full value for the other two. It is then readjusted on the lead
vessel to the full value after it is regenerated. (If it is
necessary to change the end—of—batch setting, set the full value
oneach vessel and immediatelyafter start—upmanuaily give an
end—of—batch signal for the iead vessel.)
62
-------
3. Turn well pump lever switch to ON and set pump switch to
automatic.
4. Turn controller ON. Annunciator will show each vessel in
stand—by.
5. Push sequence selector button.
6. Push controller button. Annunciator lights will show
status of each vessel. Pumps will start in five minutes.
If vessels will now yn mewn Qaogmua de dYdoj unehmwmon has
occurred. Correct alaim condition as indicated on computer
print—out and push alarm reset button. Then push
Controller button again.
7. Turn nitrate analyzer ON.
8. Check recorders for paper and ink in pens.
To turn plant off, shut off power to pumps. Leave
controller ON. Annunciator will show current status of
vessels. Shut analyzer OFF. To restart plant if
annunciator shows t o vessels in service, restore power to
pumps and turn analyzer ON. If all three vessels are in
stand—Dy, start at Step I above.
Operational Seguence
The program controlling automatic operation consists of the
following sequence of operations (with vessels Nos. 1 and 2 in
service and No. 3 in stand—by as examples):
1. A signal from the end—of—batch meter on vessel No. 1 is
received by the controller.
2. Vessel Nc. 1 is taken out of service and vessel No. 3 is
put in service. Vessel o. 7 remains in service and at
this time is one—half exhausted.
3. After a one minute delay brine and drain valves are opened
on vessel No. 1. anne pump starts pumping saturated
brine. Water supply for diluting brine is turned ON.
4. Water supply for diluting brine continues to flow for five
minutes to rinse brine from piping.
S. Slow rinse valves are opened on vessel No. 1. Vessel is
63
-------
rir.sed for 53 minutes. Slow rinse valves are closed.
6. The backwash discharge valve is opened on vessel No. 1.
7. 1)eclassificatiOn valves are alternately opened and closed
to complete declassification.
6. The backwash discharge valve is cloced on vessel No. 1.
9. Vessol No. I. remains in stand—by until the end—of—batch
signal is received from vessel No. 2.
The above sequence is repeated for regenerating vessel No. 2
and No. 3 wnen the appropriate end—of—batch signals are
received b ’ the controller.
Alarm Conditions
The following alarm conditions will, cause plant shut—down:
1. Nitrate levels in the blended water exceeding 45 mg/i as
nitrate. This alaim level is set by the adjustment on the
recorder.
2. ELectrical conductivity of the treated water exceeding the
set value. This value is set at the recorder and is set at
approximateLy 10% above the normal value.
3. Power failure. This condition was added because there is
no way to automatically restart the controller at the same
place in the operational sequence after power is again
restored. Consequently . the plant can only be restarted b’j
the operator. The operator car, determine the last
prograrc ied operation by inspection of the plant and by
reviewing tne pri c—outs which are produced prior to power
failure and after power is restored.
Computer print—outs of the plant conditions and the cause for
the alarm are given immediately after tte plant is shut—down
for the first two reasons. The print—out after power loss is
not given until power is restored.
Other Shut DOWfl gflditi0nS
Other shut—dnwfl . onditioflS occur when the pressure to the
distribution system exceeds the set value (approxi’nately 65
psi) and when the pressure at the discharge of the booster pump
exceeds 100 psi.
64
-------
Computer Print-Out Reports
The microprocessor gives five types of reports for plant
monitoring arid for research purposes. The report types are:
1. Dat jj 2Q . This report is shown in Figure 26. One is
automatically produced each nidnight and totals the flows
from the previous midnight.
2. Manual Re fl. A sample is given in Figure 27. This
report is obtained at any time by the operator by
activating a switch on the control panel. The report
contains the same data as the Daily Report except flows are
totaled from midr.ight to the time of the report.
3. j j ation Report . A sample is given in Figure 28. This
report gives detailed flow and conductivity of waste brine.
This report assures the operator that a vessel was properly
regenerated. This report will also be useful for brine
reuse studies.
4. Variables Re 22 j . A sample is given in Figure 29. The
control and reporting variables are listed which can be
entered into the microprocessor.
5. Alarm Reeort. A sample is given in Figures 30 and 31. The
information is the same as the Manual Report except the
condition of the alarm is indicated.
Chart Recordings
Twenty—four hour chart recordings are made of nitrate levels in
delivered water and electrical conducitivities of treated and
well water. Examples are shown in Figures 32 and 33.
Weekly chart recordings of instantaneous flow rates of water
delivered to the distribution system are also node at the Water
Company office. An example is given in Figure 34.
OPERATIONAL PROBLEMS
The problems associated with the operation of the plant from
the start—up period were due to pipe breaks, minor plumbing
leaks, gasket leaks, and some of electromechanical nature. In
general, none were serious to disable the plant for lengthy
time periods and most could be repaired Oy Water Company
personnel. The major problems experienced during this early
period of operation involved the ion chromatograph system used
for monitoring the nitrate levels in the distributed water.
65
-------
MCFARLAND. MUTUAL WATER COMPANY
WELt. NO. 2 — NITRATE PLANT
1 ,1/ 29 / .. S
0: 0
• EVENt: DAILY
• UNIT-StATUS .• .
• UNIT 1: SERY . UNIT 2: SBY UNIT3: SERY
TOTAL TREATED FLOW = 593500 GAL
TOTAL ‘WASTE ’ WATER FLOW 13007 GAL
äACRNASH : 3411 GAL ’ .
• SLOW. P INSE 3 8747 GAL .
coNCENT RATE BRINE : 340 GAL
DILUTE BRINE; .849 GAL
- çONDUQrZVITV .MILLINHOS1CMX 1.00
WEL,L a . 56
: TREA1tD 69
• TOTAL. FLOW INTO CITY SYSTEtI c 599400 GAL :•
PERCEI T.BL 1D n • 54• ..
• QONDUCtIVITY GLARN $ET .AT.: .
BATCH MULT. SET AT .1659 X1000 GAL
CONCEnTRATE BRINE QUANTITY SET AT 180 GAL
.0 *0’.’ CYCLES n,” 5 . SOV*07 CYCLES — • • • ‘
NITRATE ‘ALARM SET AT • ‘45 MG/L •
END OF, REPORT • • . . • • • .
• . ‘. FIGURE,’ 26. DAILY REPORT
66,
-------
F / EL( ND NUTLL L N TEE CCF4NY
WELL NO. 2 . NITRATE PLANT
ii , 27/. 54
15: 42
F_VENT: ft NL L
UNIT ST TL’S
UNIT 1: £EF:V UNIT 2: 9t Y UNIT : SEF:V
TOTAL Tr E TEc FLCW 27400 CEL
TOT L T :;T : 1:L( = G .L
I 2 ’/
SL.ON F: I NSE : 7557
:10 0
j JT. T$ :I:• E . 44 fl:
I I TYN ILLI MHOS EN>: 100
WELL
T0TL L FL0 1 I NT J CIT.’ SYSTEN
FE :C2ENT E:LEND = 54
rooucTIvIr’( AL FM SET i T
5 TC -i T ULT. . E.ET T. >.
COHS.E;TF:ATE 0 : I ‘ E OLVVJT I TV
sc v -o• CYCLES = 5
rIITF ATE AL - F: I SET T
r:tli) CF F:EFOI T .. . .
= :o: : : CAL
55 PERCE 1T
SET AT t o CAL
SOV O7 O’ CLEI3
45 lolL
FIGURE 27. MANUAL REPORT
67
-------
MCF ?F L. E’ J,L t’J’.TC C2rF ni i 1
t’iELL NO.2 — NITRATE FL IT
REGENEFATIOr; FEFORT
12/ 2/ 34
TREATED SINCE CM = I5: ’IoC’C - ,L
UNIT NO.
t F.IPIE CYCL
TI NE ‘HASTE 00:13. X l 00/Cr. I; IC —G AL
12: 47/ 4/ 24
12: 52/ 44/
5 5/ 44/ 99
12: 52/ 712/ 1L
13: 1/ 1S&2/ 174
1:: 4/ 1854/ 130
ZLC4 F :r!E’E CYCLE
TIME I IASTE CCN2. X 1C’O/ErI; IE—GAL
C: 12/ 24 / 10’)
1 :: 17/ 238/ 13’)
1:: 22/ 110/ 130
12: 27/ 91/ 130
C: 22/ 32/ 13’
13: 3 / 80/ 13’)
1:: 42/ 69/ 180
SLOW F,INSE/REOEN = 511 GAL
BACF WASH,RECEN = 2894 GAL
E:LEMD EF:rNE P.EGEN = 2i GAL
EWE’ OF rEFORT
FIGURE 28. REGENERATION REPORT
68
-------
corio ucTivi r; ’ (L4F:M SET T 5 FEFCENT
FE CEtH ELEIU) 34
S TC HLILT SET T t ’ ‘1’)u L
C’JNDUCTIVITY L F 1 3ET ( T FE SCENT
FIITF TE iL M SET T ‘IS FE CErIT
CC:ICE: ’J1R ATE EfSINE C)LI NiIT’ ET T 130 Gc’,L
Cr’CLE3 = 3 SUV ’-OT C’r’CLE
FIGURE 29. OPERATING VARIABLES PRINT OUT
69
-------
I lcr’FI_. ! 0 MUTUAL WATER CO FANY
NELL NO. — NITRArE FLA 14T
11/ 23/ 84
14: 4
EVEI,T: IOLJER FAILU rE
UNIT STATUS
UNIT 1: SB’? UNIT SEW UNIT SE rf
TOTAL TFEATED FLOW = 2473n’) GAL
TJTAL WASTE WATER FLOW = :o:: GAL
GAC I W’ SI 1 : 0 GAL
SLOW RINSE : 2éC5 GAL
COIICErITR( TE SPINE : 130 GAL
DILUTE ERILE : 4C’3 GAL
CC? W’iCTIV I TV N ILL IMHOS/CNX 100
WELL =
TREATEO C )
TOTAL FLOtI INTO CITY SYSTEM = 2 Z)00 GAL
PERCENT ELEND = 84
CONDUCTIVITY ALARM SET AT 55 PERCENT
BATCH NULT. SET AT 1 S’? X1000 GAL
CONCENTRATE BRINE OUANTITY SET AT 130 GAL
SOV#0S CYCLES = 5 SOV -07 CYCLES = 2
NITRA 1E ALAFII SET AT 4t MG/L
END OF REF ORT
FIGURE 30. ALARM REPORT: POWER FAILURE
70
-------
NOrIZELI 1D MUTUAL WATER COMPANY
WELL NO. 2 — NITFATE FLANT
i i , 22/ 84
14: 18
EVEIP: ALARM
UNIT STATL’S
UNIT 1: SB? UNIT 2: SD? UNIT : SD’f
TOTAL TFEATED FLOW = 24250: ’ GAL
TOTAL HASTE WArLE. FLCW :o:: GAL.
BAD : C ) GAL
SLOW F.IIISE : 2625 GAL
coIc: ITbATE CRIME : 180 GAL
DILUTE E.FINE : 403 GAL
CcNc.1 :T IV IT? MiLL INHOS/CMX 100
L’ JELL =
TREATED =
4
TOTAL FLOW INTO CITY SYSTEM =
FERCENT BLEND = 84
CONE ,UCTIVITY ALARM SET AT
EArCH MULl. SET AT 1659 X
CO1 ’CDITRATE CRIME DUAI4T IT?
SOV ‘-06 CYCLE.) = 5
NITrATE ALARrI SET AT
END CF REF GET
2ThSOuO GAL
FIGURE 31. ALARM REPORT HIGH CONDUCTIVITY
71
c c . -
100’) GIiL
SET AT 130 GAL
SCVr.-07 CYCLES =
45 MG/L
-------
FIGURE 32. NITRATE LEVEL RECORDING
72
-------
FIGURE 33. t.LECTRICAL CONDUCTIVITY OF RAW AND
TREATED WATER.
73
-------
FIGURE 34. CONTINUOUS SEVEN DAY RECORD OF FLOW
RATE TO DISTRIBUTION SYSTEM
74
NOTE: 100 ON SCALE EQUALS 1000 6PM
-------
1. Piping breaks occurred in the main feed water line to the
plant causing a severe leak. This break occurred during
very hot weather and at the flanged junction between
plastic and steel pipes. The area was fully exposed to
sunlight and raised the surface temperature of the plastic
pipe, forcing it to bend and placing an eneven pressure
against the steel flange. Repairs were made using a
flexible coupling between the two types of pipe.
The main product water line was also repaired in the same
way after a leak started to occur at a similar )UflCtiOfl.
2. The lower inanway on Unit 3 developed a leak which could
have caused a serious loss of resin if it had not been
promptly found and repaired. This leak occurred when the
manway gasket imploded into the vessel, loosening the
manway and starting a leak of water and resin. This leak
was detected after the vessel had been placed in stand—by
after regeneration. At this time the pressure in the
vessel usually rises from atmospheric to system pressure.
It is believed this leak -occurred due to the loose manway
coupled with the rapid drop of internal pressure as the
vessel changes from service tc regeneration. At this time
the internal pressure can become negative when the vessel
is opened to the atmosphere and a partial vacuum is created
due to the weight of the water column in the vessel. This
would Only occur before any brine water inlet valves are
opened.
Replacement and resealing the gasket required removal of
the resin from vessel No. 3 and lowering the water level to
below the manway. This was accomplished using a suction
veturl. A one—inch pipe was placed through the top of the
vessel into the resin bed and attached to the venturi
fitting to which a one—inch flexible hose was attached.
The resin was drawn out of vessel No. 3 through the top and
into vessel No. 2. This operation required about 3 hours.
After replacing the rnanway gasket resin was drawn from
vessel No. 2 back into vessel No. 3 by the same method.
Resin levels were adjusted by visual inspection through the
site glasses. The plant was placed back into operation on
the next day.
A regular routine of inspecting the gaskets and tightening
the man.. ay bolts when the vessels were under pressure was
initiated after this experience.
3. Several small plumbing leaks have occurred throughout the
year. It is believed these are due to the newness of the
75
-------
piping and its expansion and contraction due to temperature
changes. No leaks occurred on the shop installed piping on
the vessels where piping was protected by a canopy cover.
ll leaks occurted in the field installed piping which is
directly exposed to the sun and where long manifold lines
are most effected by expansion and contraction.
4. Batch meter failures were experienced three times duriny
the year. Two failures due to defective counter action and
one due to failure of the meter to transmit signals to the
counter. The defective counters which are of the plug in
type were easily replaced. The defective transmitter was
easily repaired by an electrician and it involved a
mechanical adjustment of a cam and gear arrangement in the
flow meter mechanism. One incident cf batch meter failure
occurred while vessel No. 1 was in service causing the
service period to go 50 percent beyond the set point and
the bed was not regenerated on schedule. However, the
nitrate level in the blended effluent did not rise above 10
mg N03—N/L or 44.4 mg N03/L. This was due to the diluting
action with effluent from vessel No. 2 which was also in
service at the time. Vessel No. 2 regenerated on schedule
then vessel No. 1 regenerated immediately after. This
caused the vessels to be out of syncronization.
The batch meter was repaired and the vessels were placed
back into operation.
Had the plant continued to operate out of syncronization it
is possible nitrate levels could hdve rose about the shut-
down limit and the alarm situation would have caused a shut
down. How.ever, in this case the disorder in vessel
operation was detected by the operators’ routine inspection
prior to the alarm condition.
5. The most troublesome feature of plant operation was to
maintain the continuous operation of the ion chromatograph.
Due to this problem nitrate monitoring was supplemented by
operator analysis of nitrate using a Hach field test kit.
It should be emphasized that the ion chromatograph was
installed as an expetimental test of this type of analysis
for plant monitoring. A dual instrument was available.
One portion as dedicated to plant monitoring and the other
for research purposes.
Ion chromatography requires high pressure pumping of sample
and e],luant through very tiny capillary tubes. Pressures
up to 1300 psi are needed to pump through .07 inch diameter
tubes and finely packed ion exchange columns. Continuous
pumping caused pressure increases with subsequent leaks and
76
-------
loss of resolution and vaPiation in retention times and
eventual. degradation in ability to perform the analysis.
The suppliers of the instrument are continually improving
the method with new colu ’ns, prefilters, ‘nd pretreatment
columns to overcome these problems. However, we feel high
pressure is itself a major contributor to the eventual
degredation of the method. These comment5 are directed to
the method for continuous, on—stream plant monitoring used
on a 24—hour per day basis throughout the year. The use of
the instrument for research purposes, however, was found to
be very satisfactory.
To overcome the problem with ion chromatography a different
athod for nitrate monitoring was and is being tested. The
method employs a special low pressure column operating at
300—350 psi (as compared to 1000 or over for DIONEX
columns) and uses the same micro valves and tubing as the
IC does. Nitrate is measured by conductivity after the
sample is chemically conditioned. This method was easily
adapted to the Instrument in use and has been very
successfully operated for about two months to monitor
nitrate levels in the blended product water.
77
-------
SECTION 5
PLANT PERFORtIANCE
DAILY OPERATING RECORDS
Daily records of all flow meter readings, chemical analyses,
and pressure readings for the entire plant as well as each
individual vessel were maintained. This was required during
the manual phase of operation to determine adjustment and
restart parameters and conditions as well as to monitor and
evaluate plant performance. The daily manual recording was
continued until automatic recording of all flow and chemical
parameters was accomplished. An example of the record forms
for these daily repoits is given in Figures 35 through 40.
After automatic operation and reporting were initiated reports
were obtained from the microprocessor as described in the
previous section.
MONTHLY REPORTS
From the daily reports described above, monthly reports were
compiled in conformance with the requirements of the operating
perm .t issued by the California State Division of Health.
Tables 14 through 19 summarize the daily records for six months
of automatic operation. Table 20 summdrizes all monthly data.
CHEMICAL COMPOSITION Oi WELL AND PRODUCT WATER
During the period of operation covered by this report changes
in chemical composition of the well water occurred. Plant
operation was ‘ ot changed in any way to accommodate these
fluctuations in major anion concentrations since satisfactory
nitrate levels were being obtained in the delivered water, it
was also desireDle to maintain constant plant parameters to
evaluate effects of the changing water quality.
Table 21 lists the chemical composition of monthly samples
analyzed by a certified laboratory. Total anion, TDS and
electrical conduct vity are also listed. The salt dosages and
BV(N) values are given as ca)culated as described in the
following section. It is noted that the raw water quality
gradually improved. This is thought to be due to seasonal
factors but may also be due in part to the high usage of the
78
-------
F1cFAF’LAHD MUTUAL NATEF CO.
HITPATE REtiC VFtL FLAffl LIELL ; 2 M fh .
UFERATiOII ArID lIIiIUTEUAhCE FEPURT 1 of 3
a.
G J lo Po’ji d IJ’ i ht
Coy 1 5 I 1 i
Br1cl U o U J
:d 1, ii o c of’ t r
To T-€:’tL’J L —
£ I ;: Fiscd
1
3
4
5
£
7
10
ii
12
13
14
L
16
17
I -
a .s
19
20
21
22
24
25
2sS
27
C.
3U
•IUTRLS
REt AF }. .
FIGURE 35. DAILY OPERATING t ATA FORMAT NO. I
79
-------
McFAFLAF1LI MUTUAL WATER 00.
NITRATE REMOVAL PLANT WELL #2 Month Ye r
OPERATION AND MAINTEi1ANCE PEPURT 2 of 3
u3 1 Ic iz. of LJ :te Water’ Tota l
Gallon:
D ltd c SIc’.i DacFwazh Fast
Date Br i ne Pi risc i¼in:c- Water
I
2
3
4
5
L I
7
S
9
10
11
12
1-’
14
15
16
1 ?
10
S ¼ - i
19
20
21
22
23
24
26
27
c -0
29
20
31
TOTALS
REI1AP} 3
FIGURE 36. DAILY OPERATING DATA FORMAT NO. 2
80
-------
McFARLAND MUTIJAL WATER CO.
NITRATE PELIOVAL PLANT hELL #2 I lorith . V r .
OPERATION AND I IAINrU IANCE REPORT of 3
hIltr3tC Mil1i’;r3rrs Per L.
Ikil Trc jt d Llcnd To
Date E s I Z S ystem
1
2
F )
‘ - I
4
5
b
7
9
10
11
12
13
14
15
16
1?
10
19
20
21
c_c -
23
F ) ’
C -
25
26
2?
‘,r
-F
JkJ
F )
S
TOTALS
PEMAFLS
Hours
Pt ‘ c z.
OPcr.ate
Fla i c it a i c
FIGURE 37 DAILY OPERATING DATA FORMAT NO. 3
8 ]
-------
DAILY 0PErATIMO LOG
tic FARLAIiD (jUTIJAL UIATEP GO.
D te
DALY FLOWS
lI d 1 temP C
Er i ne t ic IL
A ir teh2 £
t o da r tini’i
FT - I O U
Fr—IOU
OaUur,a to
cii tr ibut ion
lit m E tcr nc hr 11 1*2
FT—201 ne t da
FT—201 i:}- 1 is d
Gals slow rinse
FT—4D1 ne.-’t d s
FT401 this rJ3 l
031 5 b3ckwash
FIGURE 38. DAILY RECORD OF METER READING FORMAT
vic;t dw 00
tins da3 00
o r_ i
#2 meter ne/t day...
#2 tr r this d39
#2 trEated t2st r
•
FT321 ir&i.t da
FT—321 this dJ
Gal tons cur hr i ne
WeiOht ; br in?
L 7bor
hours+ t n t nuti!s
UP.Eration .
flat ntenancc
Describe tails below
Units fat rinEed
Gala Per rinse
Elec rI ne.-t d
Elec in this d
Kwhr a used
IjajE tre ted 00
OBis distributed 00
G Js wast d
Founds salt
82
-------
DiiII_T u:rc,
tI:FijFLF 3 iLl i_J iL Mii E.P rj.
WELL 1 2
P1 P2 P3—i P3—2 P2—2
hATE
PG P4—i P4—2 P4—2
TIME
P1 P2 P3—i P3—2
I I I I
l Fi_’ri F———— —— ’.!—.;C———————Lu I:1: r I——F
f ii cc I I I
I I_i i•i i t I_I I I I t i_ I i•i 1 t.
V—4C10 —41j4 . 1 2 .3
I . I I I
to I I . I I I
I -—--,, .—.IC5—- —-——— - 1 —--f—
di.Etr ibtio i I I I I
I P4—i P4—2 P4-3
F r ’.’I ,3t it: I
P5
Ft o’.i r at to di tr ib’.’ : 1 1 P in
F 1 ow r tt th u it i 1 • gp II: LI hI t 2 P i i - i t #3 . P w
FIGURE 39. DAILY PRESSURE READINGS FORMAT
83
-------
CHErI 1SIPY ‘ [ TUFT
MCFAF LIiIILI III_IrIJdL HATER C C’. :‘ c’c
NELL ;2
C o i l’S TITUE I T rr’Vl PAil TPEF ITED ELEI4D OTHEP
r 1 LI.
.3 irk L’...
( nc c c 1
CorF’cc it’:-
r-, .
ci DF ‘.1’. ¼ — ‘1
Iiiti jt’.
SI.’ 113tE
C h cr i dc
Li ’:. ,rI:c’19tc
EU
P H
- .__, — I I )
‘ —Li ‘, t l’t’’_. 1,’
OTHEP
A N’.El H.zt,r F Cc Brine C Li i E’ri’i-,e I i TP Efflin,t
E Other
llctc- 1 S rr,r’ e l1o:ri,,:i d jr .,j Z , i=24 ho_v , F ’aw ccc
FIGURE 40. CHEMISTRY REPORT FORMAT
84
-------
Tf’ )LE 11
riio: p1 (15ffl5 31i;: 1984
G 1 1on of 1:/ter / .‘r ’ 1 n Gallons uf Yaste ‘:ute : totji l
To t o .itcd I:j— ol LJ i lj I i lutv 510, C ii ii
Date — 5ysten fly I X i’acscd fine 9u i1c nj /I — I i Ire 7 inc _______ 9 ic 1 ?sn ’,v1:’ 11cr
1 177000 1/0000 7 ij OD 223 21 1 1620 2010 2313 5543
.1 17)fli ’O 161)7 (1 171111 07 1 2 10 8 3P ’9 23’) 7 ,1( 1
S . )1,’thJO 19 1380 bZlUO 411) 2 7 18 i l’) 1 1950 I S]? ))
S /7 2 1 / I l) I / 1)1)11 — 3 )t i 7,’ 9 ‘ r’) 9821) 58 ’) 101,111
/ 3080 39 8110 — 114 1’) 3 3 o) 3’ i3 2 i 9, /11
3 2 (j9cUO 20 ( lOll — 3 ”l 22 9 3jlO 114/0 /.79 223)0
9 - - - - - - - - -
10 - - - - - - - -
II 323000 323000 - 175 18 1 31 )0 7323 00 1’ 020
12 2’?SOlO 248400 6(00 203 20 0 1650 31711 23 i0 77 no
13 321800 1h, 00 5358’) 365 111 9 333)) 73 ”)) 0670 151.0
11 270800 221300 53700 176 17 9 lOt) 3670 2270 .95 9
13 IcOOuO 125500 73500 1 88 21 6 2070 3710 2? 0 9033
10 - - - - - - - -
17 - - - - - - - - -
63000 54800 13200 - 30 1 - -
19 349033 283200 05930 376 22 ‘3 70.3 1(802
20 - - - - - - -
21 - - - - - - -
22 - — — — - -
23 - - - - - - -
2’. - - - - - - -
25 /&uoO 63500 15530 - 22 4 - - - -
26 272,3)0 221500 50583 286 20 4 1760 351) 2270 710
27 161000 130500 30530 192 20 0 1600 3(23 2360 7723
28 - - - - - - - - -
29 213000 173300 39100 185 15 9 1650 3(60 2530 79 0
00 - - - - - - - -
31 - - — - - - -
TOT6LS 5 O7000 3515700 1791300 4044 vg 2! 6 3/670 8399’) 17)002
-------
1. LE 15
1 1 ,1 7 7 [ ( li in 11 /i 39 1
G I1o’n of i’ t’r Cii1’ s of V lste ydtcr
I , I — I n ii ’ 1 ) 1 111 1,1 .t si — — — N ’ L.
One __Syt:tn__ B , n I I - . 3’’ i ‘ , .e dc ‘_____nt_ 1 ,
2 29:03) 3. 3709 93 J0 3 1 . 3220 7233 ‘.(Cj 351 10
6 - — — — - - - -
& 1’22’lO .(5 3 S 520 - 2’ 3 - - - -
a - - - - - - - - -
9 - — - - . - - -
10 33i,19 1I7 ’t3 7 /3 I l i i :.) 7/ 3 7
II 21 1.3 in. I V U 2’ / 3. : I 2
I I 2. i i ‘1 31 I . 2 . :3 1 ,
13 i kO 1. _in) 26c03 1.2 2,1 ,in l 3./ 5 3 33
ft c-’’ 7/3) 43 /) 3 I 1 3 :i . -‘ “‘‘ in
1/ 2.3 2. ‘ 1 ) .1’ 3—3 1: 3 3: ) 7 ’
ia H AM •,‘t,3 1....,) 212 ‘ U i .s 3 , 7’.J
,) 3;::2 P2 6 33 - 16 . 1 I / I 3 1 2.. - ‘
23 3 1.3/J 2 ‘)liC 5 .ini 2e n 3 in L .3 61 3 1 1. 3.3
22 - — - - — - - . -
21 3723) )rz2o 4.3 1 2’ Zr 3 3.2) 0’3 m i ni 1.’I .9
- - - - - - - - -
7’, - — - - - - - - -
. 2/ - - - - - - - - -
2) . - - - - - - -
3) 2’.2’0 2233) .-‘ in, : ‘1 I. ) :3 - 2 —
31 3 3/ : ) 2659 5.: .3 —in in 5 3 , 7-in 1 :2:
TOTALS )5)in03 7 769 923T..O 3201 ., a 2: 3 ;::;i in .3
-------
it.tLE 5
PLA:T r’cI iS 1 95T 1)81
— Gdflont of dater ‘ /!
0 1 .tat ’J i/ • C’ 1 flIJ
pi le S,tte,n 6 / I /___Pas ‘ /3 ’ Ir’a’,,’
. Scuo 2.5 i i i
1 : 9 3 0 4)1 2. 7
2 , 500 35! 22 9
2 4090
3 0055 9
25t 1C0
2 ) 5. 90
21” 100
21510
1l i 30
221a00
I )r130u
21/900
16) O0
i ii 11 9
115 P ’ S
: 91 20 6
l’i3 3! 2
l9 2 / I
15500
1500
/ 10)0
33/1)
43700
30100
42100
13600
cc
-0
1 ,15’j’)IJ
2 3.53)0
3 2701a()0
4 -
5 -
6 292 0 ,0
7 , 5 )i hI0
8 /1 7)1 1 1)
9 /2 2 0 1 .. )
10 265030
½ -
13 2/00)0
I ; 291000
is iuiooo
15 -
17
13
19
20
21
22
23
25
26
27
23
29
30
31
196
310
25 6
2 9
I l 9
Q _I o’, cf . ,‘;te ‘a.c T a 1
U ’ a) L SI C i’ In’
/ 1 1 ‘ ‘ c P’” ________ 8: 1 , ‘ I a;t ‘cr
3 L i . ’ ’
3 a
130 SIC) 5 -20 1 .33
.00 37 /3 250 673)
370 1:) 79.3 60F)
5.0 ( . Y 0 c ’q
3’V’ 33 :0 Ha £255
:90 3”20 29 . 35 6 ”0
L.33
n s a v : 3 s:’a 127 a
450 372’) 2 02r (Y 1
7500 8 — 0 15020
114000 111600 2400 3/5 2511
TOTALS 3002000 2616700
335300 3272 Avg 24 7
7710 62340 47640 117690
-------
TAjIE 17
2 — — —
5 4(000 12700 0300
6 240OuO 23i400 106(10
7 195000 121300 73760
B - - -
9 - - -
13 22(1200 2 /0200 -
11 271 .000 254600 2100
12 226000 2(5100 10’300
13 23 .0fl0 213620 24 30
(3 251000 2)31.00 25 0J
16 - - -
15 - - -
17 22.000 218700 5300
13 231003 210 00 20600
19 233000 136700 ‘11300
23 317330 222100 9 )903
21 23)000 196800 65200
22 - - -
23 - - —
2 267300 181600 82’OO
25 232000 193200 83000
26 200 176500 71503
27 36000 16/903 168100
28 15500 15/500 -
29 - - -
3 : — - —
31 — — —
239 2 4
215 21 3
‘ 103 27 0
- 21 7
353 3 I
P1.AIT 11 C0l’L ’S SI ‘1 [ P (0.11
Cj11on of ‘1jter Av ‘j Ca11 i’s cf Paste ( te Total
10 1 .e it .d fly- of t’ .iij Dilu o SIc ’
/Pa eJG i.e — 1111 tL .1 ;f1 1. e ________ B.c. 1cj ir_
a,
116 20 7
2’6 15 5
214 21 1
403 25 1
113 24 4
415 21 9
182 25
167 70 0
43! 26 1
177 26 9
183 26 1
155 24 0
178 22 8
3130 2611)
— 3Y1 - 0. —— 2730
3620 2/20
7120 5350
3//0 2613
6953 626’)
31.50 2 10
3550 2 G0
( ( ‘/0
1 1 )3 1)
1) 0
2120
1010
2 70
1060
1010
20 3
970
1610
990
960
1050
1070
1830
1)10
6830
7530
7380
5u 7Q
(“31.0
703)
7350
I 3933
7200
1130
£370
7270
4)20
2 70
2720
253)
2650
£2 .0
3101)
3 ,I’o
32j .
3553
361.0
j ’ 13Q
14 .30
5500
2360 7160
2)0
5(43 1 3/0
5140 1’010
iOTILS 4244700 3433100 811500
43u7 Avg 23 4 22300 773 ’.O
57350
-------
T 9L1 0
P1_A 11 1 LOitIS 00 (ic) I ) I 902
Gafluic of Watir a Cc l i c’ , of Wastp 0 , ctcc rota
To 1 t ci fly- Ta I y Ic u (a I - Ga a:
Date SySlL lc3J i 3 I ’d td !3rirc H ’ ., ji J/] )), I L Pin Oar) ht: (er
I IOlinO 1)1100 — 185 20 8 1071) 35 (0 2 6 ’2) 1020
2 3jo.o : IVEOIj I 19830 104 7 9 970 3 U3 2( 00 6910
3 2 39019 1 ‘6100 I 12010 — / 6 — — — —
I ?7lit9) I 7Iuo 10i —(n) 313 31 - - H ( 1, 3 1(0
5 25)uutj i i 1110 0 It (JO 2)2 22 0 990 3 .9tJ 2100 71
0 - - - - - - - - -
8 255000 100300 95700 19: 23 4 990 33 0 ?YO £923
9 202000 130380 72700 333 22 9 1690 6920 51(3 14(90
10 179 1)t0 112800 61000 - ‘:5 6 - - - -
11 ‘ :00O 21600 1Ca 00 - 19 4 70 40 - 133
12 236000 153000 81000 374 ‘2 4 ISCO 49) 5250 iC 3
13 - - - - - - - - -
14 — — — — — — — — —
0 0 15 255000 163201) 119700 - 23 6 - - -
18 232000 132100 79700 139 2J 3 1420 101)1 1925 220.1
17 2c9(I1)0 14)300 7550J 206 23 6 92,) 3Hj O 1013 57 10
13 190000 °330 5]/P9 455 25 9 133 3130 75?) p710
19 2210J0 142000 79000 180 23 7 97) 3520 2970
20 - - - — - — - - -
22 - - - - - - - - -
22 235080 167600 074 160 20 7 880 3590 2720 71 o
23 15601)0 99200 5(8 - Ic) 1 - - -
24 2C3c00 180300 1030 125 20 2 10 0 3320 13.0
d5 21201.3 173)00 082 280 21 3 251 3991 2710 61 3
26 46200,3 272300 18) 359 21 5 790 7180 7.0 22310
27 - - - - - - - -
28 - - - - - - - -
29 21000 6300 14700 1 22 7 - - 190 i0
31) - - - - - - - -
31 - - - • - - - - -
TOTALs 4738:00 3054600 1583800 3752 23 3 16620 57360 451 0 3013
-------
1_I. 3
P I ‘0
Ct ‘ite r — —
— ‘C ) I: - — E
C C .t L 19 / Y
‘,i’.te it_”r
(3c.
23 I I
r 1 ;; of
‘ ).
‘ ‘ -
t i
rj . ,
3
.
-
S
.
-
C
.
-
-
3
-
-
-
9
-
‘0
.
-
j I
—
13
•
—
—
—
—
•
-
: .
3,
‘5
3’
C_a)
253 )
; ‘‘jh’)
‘1
‘2_ 3O
J ;
•.,
c’7u U
1;
jr;
1 12
—
.
z l
—
1’ )
7
23
—
— i i
::,
J ”
—
If ;)
..
IC 3
—
63 s0
( 7 ,
:9
23
/1
:5
-
15, 13
13’0iO
-
JCU’)
2-.02’J
-
1810
201
—
;3
2! 7
-
28
233
-
153 3
-
U _3
Z920
—
2952
5 ( 3
-
I I 3o
2
lIuiOO
4033
2 1)
21 5
313
3 !o 0
1953
h I S
25
:t
27
23
29
33
31
-
I ”’ I
3 1 ; ! 7
7325)
559 U i
6510 :3
—
-
I / ’ 3
3230
6)’ 3c)
iF 0
552 933
—
-
21530
‘ifl)
L1l’O
lIcy ’
108100
—
.
(U
35)
29
7’S
!3
—
-
, 3
3
23 5
23 1
35 0
—
-
120
03 -)
iflø
2373
5 10
—
7210
72 19
1? ‘
l. , 3 0
j 3 3’
—
7313
227 )
; s 3
13 73
2233
—
roo
1t(3
:
I 7y o
1: o
—
T3 iS 317220’) 3262530 603520 4022
23.? 1 193 71273
3 )4 43 115513
-------
TABLE 20
SU 1MARY OF MONTHLY DATA
*
S j _p o _ ______ * *
Lbs/ 1000 Gal 7 Treated Nitr ite mg/i 1000 Gal
Month Lbs/Cu Ft of Blend n Blend n Blend Del ivered
6-84 5.94 2.01 66.2 21.2 5,307
7-84 6.36 2.42 74.0 22.3 3,595
8-84 6.46 2.69 87.2 24.7 3,002
9-84 6.48 2.69 80.9 23.4 4,25
10-84 6.35 2.10 64.5 23.3 4,738
11-84 6.55 2.82 83.8 23.2 3,771
, verages 6.J6 2.49 76.1 23.2 4,110
* t’lonthly averages
91
-------
TABLE 21
CHANGES IN WELL WATER CO POSTTION OVER ‘11 MO iHS
Month Nitrate Sulfate Bicarbonate cnlor ide TDS Jot 1 Anion BV(r Salt
1-84 71 115 104 C8 840 537 7.72 254 5.33
4-84 65 95 113 84 900 590 7.25 2Q5 4.90
5-84 60 100 100 77 730 490 6.86 286 4.59
6-84 56 95 86 75 *ft4) 413 6.40 298 4.25
7-84 58 85 70 74 613 335 5.94 30 4.13
8-84 50 80 69 68 600 397 5.52 331 3.63
9-84 49 76 75 58 590 390 5.24 360 3.47
10-84 51 60 78 60 590 370 5.04 365 3.37
11-84 62 80 82 62 *( A) 350 5.76 307 4.19
* NA = Not Available
** Lbs NaCi/Cu Ft to give 30 pp ii nitrate leaL age at 100 bed utilization.
-------
well made possible by the nitrate plant.
The plant was set in January 1984 to treat 260 bed volumes of
water per eacn service cycle and a brine dosage of
approx nate1y 5.61 lbs/cu ft (1.50 LW of 6%) to accommodate the
water analysis Lot that Month althuugh that brine dosage was
not always obtained as icen in the records given below. The
change in water quality demanded less regenerant and longer bed
life for optimum performance as shown in Figure 41.
Consequently, It was assumed that process inefficiency would
result from holding bed life at 260 13V and salt dosages around
the 5.61 value and that the loss of efficiency could be
estimated and cc mpared to actual data. The data on TDS and
conductivity were obtained to sec if these easily measured
parameters culd be used to estimate bed life and salt dosage
for maximum efficiency.
Table 22 Lists the monthly data on wa.er compositions of raw,
treated and blended water from certified laboratory analysis.
The January treated water nitrate (26 mg/I, agrees with the
leakage estimate of Figure 41 (28 mg/i). The leakages are seen
to drop as water quality improved. Nitrate leaL.ge values are
discuEsed the follo Lng section.
EVALUATION OF PRLMAR’ PLANT PERFORMANCE
For tne purpose of this analysis plant performance data is
considered to be primary and secondary in nature. Prinary
performance relates to chemical factors involved in the process
such as nitrate L’ aKacje , regenera.it used, brine efficiencies
and effluent histories. These factors together with the resin
used determine the primary adjustments which must be made to
operate the plant successfully and efficiently. These
parameters are also predictable based on various models of
resin interaction with feed and product waters.
Secondary performance factors relate to non-critical plant
adjustments such as the amount of water used for diluting
brine, blending raw with treated water, and backwash ng. Tl sc
factors, although important for proper plant operation, are not
critical for achieving best cnemica]. efficiency but are more
related to hydraulic efficiency. These parameters are more or
less left to the operator to detarnine and may vary due to
physical design of the plant.
93
-------
tIUILE 2?
MONTHlY P N1ON ANALYSES CC’ T1FiE0 LACORATORY
IIG/L
* 0 1es than 5
+ Concentration of nitrate hstcd as ug N03/L.
To convert to g N03—N/L div de by 4.43
cat otiate
oi t.h
1-34
4-84
5-84
6-34
7-84
8—3
9-SI
10-3
11-34
Ch I or
raw trt bind
raw trt b1, d
71 26 60 115 0 31
66 21 37 95 0 33
6 20 28 03 0 16
56 13 26 95 0 27
58 14 24 85 0 21
50 10 14 30 0 0
49 3 20 7t 0 17
51 1 31 C 0 36
62 17 23 S 0 i 23
ra’,
104
113
100
70
69
80
82
trt
100
21
38
75
57
61
13
?1
17
bind
102
38
80
50
67
55
49
50
5 1 •
83
203
84
208
77
226
7b
177
74
174
53
155
58
143
60
165
62
159
169
161
177
158
148
154
128
112
121
94
-------
CRITERIA OF PRIMARY PERFORMANCE
Primary performance data is evaluated by comparing the actual
performance data with that which can be estimated from ion
exchange theory. The major parameters of interest here are
nitrate leakages and the consumption of regenerant salt per
amount nitrate removed from raw water. During this period of
operation only one brine dosage (although brine dosages varied
due to lack of precision control) and bed life were used while
blending percentages were frequently altered. The latter
parameter is not considered an adjustment to the ion exchange
system per se but rather an adjustment to the distribution
system only because changes in blend percentages do not effect
brine efficiency (BUF).
Three criteria can be used to evaluate plant performance.
These are:
1. Salt Dosage Requirements
2. Brine Use Factors
.3. Nitrate Leakages
4. Effluent Histories
These quantities can be estimated from theoretical
considerations and compared to actual performance.
CSTI 4ATES OF SALT DOSAGE, BRINE
USE FACTOR, AND NITRATC LEAKAGE
The amount of salt required for regenerating a nitrate spent
bed can be estimated by the method described in Reference I.
provided a Type I or II strong base anion resin is used.
Various factors must be known or closely estimated about the
feed water and product water compositions and resin properties.
Actually, before attempting operation of the plant such a
calculation should be made to adjust initial salt loading and
useful bed volumes to breakthrougn, BV(N).
From the chemical analysis of Well 2 water obtained in January
1984 and using 1300 meg/I for the resin anion capacity and the
value of 4 for the nitrate to chloride selectivity constant,
K(N,C) the data in Figure 41 was obtained. The Li.rst four
lines represent the known feed water composition. The fifth
and seventh line are the resin properties of the strong base
anion exchange resin. The sixth line is the calculated amount
of BV(N) for the completely regenerated resin. The eighth line
is the nitrate leakage which is desired in the initial portion
of treated water. The remainder of the data are calculated
from the above inputs.
95
-------
NITRATE P16iL. .= 7F
SULFATE MG/I = us
CHLOR IDE ME.t 98
•BILARB&pfA IIGiL. 104
YOL.CAPAC ITY (EO/L}a 1.3
BY(NYa 322
NITRATE-To ck [ oRI EQCONST
N03 MG/L IN TREATED WATER -
MOLE FRACTION 1103 IN TREATED WAlER
MOLE FRACTION fl03 ON RESIN- AT RUN START
BY (N) ‘OF-- GENCRATEfl RESIN
POUNDS NAC!. TO REMOVE 504 FROM -0)- IS Cu i ’t
MOLE F ACT. N0 o r RESIN END OFRUM
POUNDS NACL/CJJ- FT TO REMOVE N05
•P2UNDS NACL Ctf FT .REthN IJEEE.S
• BY OF SZ NACL NEEDED
BRINE USE FAcTe ’
POUNDS. NACL/100G GAL, TPFATED WATER
I ER CENT TREA 1E!) 114 BLEND
POUNDS NACLI 1000 GAL BLENDED WATER $ 30
a .0524628374
= .199956 444-.
= 258.055753
RESIN= .2.25340203
= . 3fl629065-
3.44948?45-:
— 5.702 Sa1s
= 1.524W6’41
-
2.95447Q87
= 95.-3488t72
IIG/L N03* 2.2170flo2
LiIS1TER ikt D
FOR t00 .
CCLU?!; CH t I S n R’f -
E’ Z tiT LILt r:-Ju -
t. 4
- - - FIGURE 4.-WAtER AND COLUMN CHEMISTRY
- - - - JANUARY -1984 ( OO%)• - - -
-- 96 -
I -
-------
Using these estimate the plant was ad3usted in January, 1984,
to use 1.5 13V of 6% salt as regenerant and 165,900 gallons for
each service cycle or 260 13V(N) to give an approximate nitrate
leakage of 28 mg/I.
In November 1984 the water quality at yell 2 had gradually
changed to that shown in Table 21. A second set of
calculations based on the November water quality are given in
Figure 42. The calculations show that if salt dosage were kept
at the same level as above a longer BV(N) can be obtained as
well as more efficient Brine Use Factor, BUF (meq regenerant
per rneq of nitrate removed) . Although the w ”er quality of
Well 2 improved during this period no adjustments were made to
increase the volume of water treated per service cycle.
Consequently, lower BUF factors were not realized. Instead,
blending was varied to take advantage of the improved water
quality. This could bt done on a day to day basis without
d sturb1ng the plant operation. However, as seen below,
effiriency was sacrificed. From Figure 42 the November water
at 23 mg/I initial nitrate leakage 329 bed volumes could be
treated instead of the 260 for which the plant was adjusted at
the salt load being used (1.5 13V of 6%). Thus only about
(260/329)’clOO or 79% of the bed was actually in use by ending
the service cycle at 260 bed volumes. Instead of obtaining a
brine use factor of 7.15 a.s indicated in Figure 42 this factor
would be 7.15 x 329/260 9.05.
Consequently, one would expect I3UF to be about the same as it
was in January and to remain in the B to 9 range. The actual
BUF values obtained are given in Figure 42 on a month to month
basis and are seen to average 10.3 throughout the six month
period.
One of the important conclusions which can be drawn from this
analysis is that salt usage can be decreased by allowing the
length of the service cyc1 ’ to adjust to either incoming water
quality or to nitrate breakthrough. This would require
frequent manual adjustment or automation sensitive to chemical
analysis. In this case a salt savings of 100x 8.48—7.06)/8.48
or 17% could be realized based on this calculation. This
conclusion will be tested in future operation.
The data of Table 23 summarizes the brine usage data obtained
over the six month period of operation as calculated from the
monthly records of Tables 14 through 19 and TabLe 22. It is
noted that salt dosages were higher than the target value of
1.5 BV of 6% brine. This was due to the inability of the brine
system to deliver a reliable consistent brine dosage. The
system relied on a timer control1 ng the run length of a
97
-------
•N T ND COL’ r. o: E i 1 5 TRy
CR i : :. BE3 UTILIZATION
NITrTE NEiL 52
SL’L - TE ML ,L
CHLORIDE MG/U
E ICAR ONATE MG/L 32
VOL.CAF’ACITY -(EO/L) 1. .
EV (N) =
NITF:ATE—TO—CHLO ;: :5 Eu oc:
NOT MG/U IN TREATED WATER
MJLS n .CT CH iN i TFL t hM1 -‘
MOLE FRACT ION ON RES:N AT RUN START
D V (N) ‘OF PEOEME .ATED F:ES IN
POUNDS NACU T2 REMOVE £04 FROM ONE
MOLE FRACT. NO CM RES 1 N ENO EFRUri
FOUNDS NA:L /c j ro REMovE i :co
F’OUNLSNACL/CU RESIN NEEDED
By QF MACU NESDED
FOUNDS I- CL/ GAL FJ:TATED
p CENT TREA Si)
FOUNDS NACL/ iooo ELENEED
tiA 1 ER
WATEF:
—
=
=
2. 27 ’1i)OO
= .
= 5. 545O i ’
1 . 45257524
= 2.
L:. ii; :
riG/U ;i2 . E44 42:31
FIGURE 42 WATER AND COLUMN CHEMISTRY
NOVEMBER 19.84 (IOOd/ )
CU FT F:ESIN=
981
-------
TABLE 23
t O JTHL? BRI E EFFICIEt CJES
Salt_Dose _____
Brine Use Factors 13V
flonth 1 3 Pveraqe Lbs/CuF 6 Brine
6-84 7.5 8.5 9 0 8 3 5 94 1 59
7-84 8.6 9.1 9.9 9.2 6.36 1.70
8-84 14.7 10.2 10 5 11.8 6.46 1.73
9-84 13.1 11.8 9.? 11.4 6.48 1.73
10-84 9.8 13.3 10 9 11 3 6.35 1.70
11-84 9.2 10.5 9.5 9 7 6.55 L75
Average 10.5 10.6 10 3 6.36 1 70
* 1 = Fro n onthly aver ges of Hach i t nitrate analysis
2 Fro n nitrate analysis oy certified lab on blended product
3 = From nit at’ analysis by certified lab on treated mmater only
Equivalents of Chloride in Fresh Re enerant
Brine Use Factor —
Equivalents of Nitrate Removed from Infi uent Water
99
-------
centrifugal brine pump. Urine rate of delivery varied due to
variations of brine and diluting water pressures. Consequently,
brine rate was ruanu4lly adjusted frequently and was also
adjusted to sligntly nigher than the gpa required as a niinimum
brine flow rate. Changes in brine control were later made by
reprogramming to terminate the batch on flow rather than time.
The estimated performances at the orine dosages listed in Taole
23 are given in Figuros 43 throujh 4 .
Table 24 provides a comparison of the . ctual chemical data
obtained over tne six month period with that which was estimated
for a plant using 2 0 by service period. Tne estimated nitrate
leakages and SUF vaLues were e. timated based on the actual brine
dosages used in the plant.
Nitrate leakages oo ained by the plant are in good agreement
with the predicted values. The actual leakages can be less than
predicted because in the partially regenerated oed nitrate
leakage gradually drops as the oed is operated due to the
depletion of nitrate al the exit end of trio oed. Also, any
failure to obtain proper declassification will give an erroneous
nitrate analysis. Trie nitrate leakages in Table 24 are from
grab samplc s for certified lab analyses and are suoject to
variations of sampling aria analjtical errors.
The plant dUF values of Taol 24 are the average monthly values
taken from Tasle 23 wnere 00th daily Hach anal ’te and rnonth y
certified lab analyses are taken into the averaging process.
The data show that the plant is about 3 crcent less efficient
than is pred.i:teu. This is quite remarkaole considering the
estimates which are made in the method uses for making these
predictions. (Reference I) One gross assumption in the model
used is that nitrate and sulfate are absorbed homogeneously
throughout the oed. In actuality nitrate concentrates at the
interface oetween the operating portion of tne bed and the
unused portion. Wherever nitrate is concentrated less
regenerant is required to remove it due to the shape of the
regeneration curve. Consequently, lower UUF values in operatiol
may De expectea than those predicted. Another way to look at
this is that sulfate absorption in the upper portion of the bed
moves the nitrate closer to the bed exit making less work for
the regenerant to remove the nitrate.
Because the bed is partially used the distribution of ions in
such a bed especially after several cycles is probably closer to
the models’ assumption of homogeneous distribution than would be
found in a fully used oed. One would expect, therefore, that
better brine efficiency would oe obtained for 100 percent bed
use than is predicted y the model.
100
-------
WATER inZ’ COLt flN C ShiSThY
FOR PLANI’ bY OP 2S0
;UP.XrE I’”j/t,
• SULFAfl rGiL . = i
CHLORIDE M5/L -
bICARPOnATE ’P1f /L ‘ -
IrIL.CARACrTV CE(t/L) i;3.
WCN ’ 3C.5
r9IIRRTE—TO-CHLCR1IE EQ CC:45T
N03 tieii.. I N TREATED . WATER
MOLE FRACTIO n ItS IN ThEATED WATER
MOLE FRACTIDN 1403 ON RESIN 0? RLk4 ST4 RT
•E VGO’OF KEfENERATED REPIN
PFJLt4DS NACL TO REMOV’E 504 FROM ONE CU FT
MOLE. FRACT, N03 -ON RflIU END .OFRUI4
POUNPS NSCI. , U) PT TO cEP OVE N03
ounDa ACL I O U FT RES IU r4EaD zb
SV OF OX !UAC.. NEEDED
BI-UNE U’3E FaCTOR
OJ2CS F4ACLflCOO iS? L 7’ ATt!D WAt fl
FER CENT TRtMTSD ‘1$ eLEkD
.RQi.!NL’S NACL/LOO’) GAL. BLEtU)ED Wtvrcft ‘l!•
z 4 • .
=17.t3
— •o44oo9ô aa
.1$5 a92 6
=- 3:34. 2462&Q
RESItI I.87t5531
•
• 4.0600$ó’-’.
- =
- =
— 67.fl24aTh
Sc P1SIL fl03 2.O61dt .
FIGURE 43. WATER AND COLUMN CHEMISTRY
- -JUNE 1984 (250 BY)
-------
ti Mith ArID CCLUI1N CHEMISTRY
FOR PLAN.r BV: CF ZoO
NITh T€ WS/t...
• SULFATE ?1S/L
• D4LORIDEMS/L 74
• BICARBONATE M5/L — 7Q
YUL.CAPA ITY CEO/L)1.3
• BVtN) 417 .
NITRATE—TO—CHLCRIDEE0 CCNST
N03 M5/L . IN .TPSATED. WATER
• MOLE FRACTION N03 IN TREATED WATER
• SLE FRACTION N03 ON RESIN AT RUN START
BY(N)’OF!.REZSNERATED RESIN
POUNDS NACL TO REMOVE 534. FROM ONE
•flOLE FRACT • P IUS OW RESIN END• .DFkUN
POUNDS NACL/CU FT TO REMOVE NOS
• POUNDS NACL.’CU FT BESIN NEEDED
• BY O 6Z NACL NEEDED
URINE USE FACTOR
FOUNDS NACL/i000 bAL TREATED WATER
PER CENT TREATED IN BLEND
- POUNDS PJACL/i000 GAL BLEPIDED WATER
4.
= 14e97
S .0406576098
n .144956701.
= 3 6 9h Z69
1.679 10669’.
= —. J,_.
• a’.
S
. 6. a29e73 ’
n 1, 7)132!4S
— 9.42932191
z.:71 7!’ 7
n
CU Ft RESIN
@ 30 M6/L NO3= 2.1’2’ Y7592
FIGURE 14. WATER AND COLUMN CHEMISTRY
JULY 1984 (260 BV)
:102
-------
ER. - iL LJL .+ ;
i I:E ML
I L: :R 4. -½ F 1 R/L T.
N I Tp TF—TO-- HLcE• IDE EL DOHEl
N0: 1C /L IN TF Et IED I TFF
MOLE FF .LTIL: ..j NO. :. IN fF:E TED WATER
MOLE FR LTjCN NOT . ON F:EEIN AT RUN ETART
E;V (N..’ •FL ERA1E: FIiRIN
F€JUNL5 NACL ru REMOVE: R04 FROM ONE OU Fr
x END
‘DUNE.:.: .r ’r TO F ENDUE
FC:Ur.. DR C ..’.U FT FLIRIN NEEDED
‘4EEDE O
EF;INE u :
IN
F’ U :.$, 1: _ EL: Da:)
4
=
FEE IU= 1.
= : 4 4’
= . iL ,’L .F..1
FIGURE. ‘ 5. WATER AND COLUMN. CHEMISTRY
AUGUST 198 4 (260 BV)
103
-------
— —NIE iC _L :
5LL:- .TE t ?/L_
:u. L. I
.E VCNi) 477
Ni IF rE—To—-c:--1_cF i EO OCN T
r iu:.i MC .’L IC tI IF 1Ei) t. T :k
MOLE i: T I T :L;TE’
VOLE. I :1; N0 ON : 3 I C H I N i i
E:.f ATEL
FF 1 - i :IIIC:
O L E Zr 1 i - O : _.:r
FL.U L l.,I:,::.J:J. ETTO - E -!1IE
Fc:LNLs r o i - s. ! i
II— —
ICI t f’:T TI
:::!rlr. N ACL_. I
TEI TEL
:. ;_ . J)
.4
= . II .
4.
I -
FIGURE 46. WATER AND COLUMN CHEMISTRY.
SEPTEMBER 19814 (260 By)
104
-------
: WATER AND coLerin C.HEM £ &TRV
tp PLANr BY OF 250
NItSATE rtâ/L
SULFATE NE/L n. Q’ .
CHLORIDE NE/L . 60 -
BICARBONAtE NE/L 78
YOL.t %#AC1TV (E( 1 )/U= 1.3
• BV.CN 521
• 1 IiTP.ATE—TO—CHLORIDE £0 CONST
NOS NEIL IN TREATED WATER
•: MOLE FRACTION Nb3 IN TREATED WATER
MOLE FRACTION NO LIN RESIN AT RUN START
BY (N) ‘OF RE ENE MTED PESIN
• POUNDS .NAçL it) PEfICVE 504 FRt fl OWt Ci.! FT
MOLE FRAth. N0 C M flSIN END O RUN
• PQUrIOS NMCLICU FT 10 R!MOY N03•
• POUN)S I’IACL/CJ ST f uN HEELED
BY .OF. o x )$ CL NESDED
!RINIi t! 3E.FAçTCS . .
• • s aAi iw-.CL/ c•cu Tra tcEtD W\rER
PEI Ct,7 1 FZ IN isL&14D.
• !0uH :4ACt.. .t’)?C! E piL SLE!IDE$ IsJftj?ER ft
.= 4.
• .1O.7
— •0342 52O144
= .fl4177678
— 4 o.esjsa8
RESIW 1.1?454 st’
.254 17767 8
—.z.16 0l17.•
aZ49557O
1 . 69774247
=
e3.2O4S697 .
•n 52.L054611
ttflL NCi I .701C0732
flGflE 47 WATER Mb COLUMN CHEMISTRY
OCTOBER1934 (260EV) .•
s. •
-------
WATER AND COLt2M; a CHEMISTRy
FOR . PLANT LV CF 260
NITRATE i i _ a .62 -
SULFATE MG/L SG
CHLORIDE MG/L ..
- BICARSbP1AYEr%/L- 82
VCL.CAPACI ’ CSti/C)’a 1.z
!V.tN)= 420 -. .• .
• NITRATE—To—CHLo.R IDE EQ CONST- 4
P103 Mt /L IN TREATED WATER • . . • . 14.6
MOLE FRACTION P103 . IN TREATED WATER . • . . — o4o9o1o23s-
MOLE FRA T ION P133 ON RESIN AT RUN START n • 14!723Zá6
BY N)’OF.REGS9zRATED RESIN. • • . . .35 .24 549
POUNDS NACL TG REMOVE S04 FROM ONE tU .Jfl RESIN— I. 79394 3.
MOLE FR Ct. P103 CN RESIN END CFRUr4 • • .•• . . 2?S62461
PQUNDS NMCL,;Crj FT TO :REMOVE -N03 • • :— 4; ,o S
POWNQS NACL/CU FT RESIN . NEEt -W • . . = 6. 492 ; 3.
9%’ OF 6Z N L It€DEU . . • . ..
Us FActuw: . . . • . —
POUNDS NACL/1000 äML P.EATV3 WKrgR • ... .. . 3afr Ei6o
PER. CENT TREMTW IN ELEND .. • *7 LCs54b
MJUPIDS ;lprL/ l000 GML SLENDED Wn!tR • r1G/L -103= .273;s4o7
FIGURE 48. WATER AND COLUMN CHEMISTRY
- •• NOVEMBER 1984(260 B )
106
-------
The net effects of regenerating a bed which is not 100% used
during the service cycle are:
1. The working portion of the bed will have more nitrate
removed by fresh brine than is predicted from the above
estimates. This, nowever, ma 1 not e a siqniticant amount
because of the small amount of resi’. in the nitrate form.
2. Some of th nitrate removed from the working portion of the
bed will be retained in the buffering portion of the bed.
This is favored by the high chloride content of the resin in
this part of the oed.
EFFECT OF UNDERUSING A BED O . BRINE EFFICIENCY
The method is availaole to allow comparisons of brine
efficiencies for fully u eo and partially usea beds. dater and
column chemistry parameters arc presented in Figeres 49 througn
54 for 100 percent uti’ized Duds. The same monthly influent
water compositions are used as were used in Figures 43 througn
48 using only cbO Dud voismes for service. A comparison of
estimates for the two different cases is thought to give a valid
estimate of orine savings whicn can se realized if plant
adjustments are continually ae to keep the bed fulLy utilized,
although, as pointed Out aOove slightly better efficiency is
ootained than predicted with the partial bed and even better
efficiency may be ootained than predictec with Cull bed use.
Table 25 summarizes the 81W values for the two different cases.
It is seen that the average estinate (or brine savings is 17.6
percent if a full oed is used. The 100 percent used bed
performances show that a sliynly higher nitrate leakage would be
obtained together with less blending percentages and less salt
requirement per 1000 gallons of blended water.
These estimates will be tested in future operations of the
McFarland plant. The major concern in using the full bed is
that close and continuous adjustment of the plant is required to
conform to water composition changes and the danger of nitrate
dumping” if an over—run should occur. This, in turn, would
require close or almost continuous nitrate monitoring.
Effluent H is tories
Breakthrough curves can be easiLy obcained from the McFarland
Plant using the sampling ports provided which penetrate into the
interior of the bed. Water samples were withdrawn from Port No.
2 which is physically about 40% below the upper surface of the
Oed (assuming 36—inch bed depth). Hydraulically, the prooe was
107
-------
N TR TE
——
_. F;-CT 1.T
NIT 4TE—TO—C:HLCR IDE EQ C ST
NOT r1i /L r u r ir TEE ti4IER
MOL. E CT I I .‘;T-
ifl —: -. 2 Tr ..T CN F: .IN r F:UN
:E I
‘OJ iL r rZ2L. 12 F:EMO .E O4 FP M
MDLE - TT. N’JT N FESL’1 O :UH
F-C J! 2 - —CL,C2 FT TO R MDVE
-Es
N iL ; O
o: D F4 2L C) TRE TED .; E :.
:::i• F.i :;
J !D j: : C— L EL •)Ei
= 4
2’.
1i1:T Ti
FIGURE 49. WATER AND COLUMN CHEMISTRY
JUNE 1984 (100%)
108
-------
r ITF TE Mt3/L =
—
CHLOF .JDF. MG/L . = 74
E ii’L. =
VOL. O: P•CiTY EO/L I.
V(N) 417
N I TR ;TE—TO_CHLUR IDE EL i LONET
NOT. M6/L IN TF E TED W TER
MOLE F t ’ t i i ION NCI i N TF:E-TE.D
MOLE FF; CT I ON NOT ON F E3 I N FrON 3TOR I
E V N) OF REC ENEO iEE IN
ifl I fl. r ’I r ; E.iO)E S:O F :Or1 C’ F LU FT
‘iQLE TF: iCT Ni:1 IN i HL
Nir:ioL/::u T TI: FEM.JVE ft ::
F:CUMES F AOL/LU Li FE N
:: Ni :_2L .I:EEEI)
FCuN22 N -iO . / :‘Y
FrF C .iT E: -T 2
FCUND5 iAOL/ I :.:::• ::. G- L E:L !DEO I.. .- .Z
19.
-
2.
.—.— .
:t-.S• 77
f !: f
=4
FIGURE 50. WATER AND COLUMN CHEMISTRY
JULY 1984 (100%)
109
-------
— ia c: :i_
NITF:( iE NS/L
SZJL —- TE YtS/L
CHLORtr)E M3,’L =
Ei: ;Fc •rE ‘3 ,
VOL. C F C TY O/L)
E V(N )= 4 S
NI TRATE—TO-—C:HLOR I CE. EQ CONE.T
NO:. NS/L IN TREOTEI) w TE :
MOLE FL. CT I flr . Nfl J i J 1 :;: i SI) N 1EF:
MOLE FFt CT I ON NJ . ON F:ES I N T FUN START
E V (N) OF FSOEriE TED I N
POL’N!DS NOOL TO F:EMOVE 80.4 FROM ONE CU FT
:oo Cr: kEEl END CFFLN
flft fl’ : I’AOL : T TO RENOOS NO::,
FOUNDS NACL/OU FT FEE IN NEEDED
o :: N.OL
bRINE NEE F cTc.E
FOUNDS N CL/ I 00) S L TREATED N TEF
FEF: CENT TF.S ZT5I) Ir.: SLEO’ID
FOUNDS Ni CL/ I oo: E L SLENDED NOTER 0’
4
I7 4
992 :’ Ec
1. ) :
. I7 3
,.
I 7.l :l
=
=
: n MS..’L :o=. .
FIGURE
51. WATER AND COLUMN
‘AUGUST 1984 (100%)
CHEMISTRY
RE:3 IN=
110
-------
NiTr TE M /L. = 4:: ’
u :r t1 /L =
C P( iT /L)
E V N) = 477
NITF TE—TO--CHLCF IDE EC CON 3T
r o: MG/L N TRE rED W :TER
rlf2t.7F:( C TIO 1 ;:H Z : : :E
MOLE FF:)’ riC HiI. ON F:C: N4 4T UI
H) ‘OF L F TE ) FE C H
F:EC . E ) C• FE:Th ç N:: CU
r- —- - —
i: :: : o.
F’ J ’!D N
o’. ;; : .-- rEF
: : ..E’ • CL/ i : :’ -‘u o 7F: :1EiL
FIGURE 52. WATER AND COLUMN CHEMiSTRY
SEPTEMBER 1984 (100%)
.1 --
—— —r
= 1 7
i7. i7
1N r 2. 57T13
= 4.
= ‘ ,. 4 .-‘5 : :’ -
-=
111
-------
• WATEé’ c CüLU ! fl f;HSM LbT¼Y
ICTRATE I IG It. • 5t.
E%JLF.Alt IIU,/L.
CHLOS!DE MG/L.
B::n ’:9A i / I ;
Vth&.’CkPIiCTV t&!WL) 1.
• B’Q N) Z20
• N1TRATE tOrcHLOR!DE .EQ CONS!
NO; MG/L tN tREATED WATER•
MCLE PRACTIQN NC3 IN nEATED WATER
flCLE FR CflON NOS 0N RESZN AT RUU START
B t s ,aJ SCF aE EFiSF:AttD AEEIN •
PCJ! iDS 1IAçL. TO $EMOVE 2134 Fi O 1’i ONE CL ! fl
MC!.! FF.aCT. P13 Oft E!1N EN! OFWJN
PXflDS rJACL/CU FT TO REl OYE S 103
• M2UM0i NACL/CU- FT F.EflN NEEDED
9M DF aX NACL i4E!t -
*ari€ L W4 .iTbR .
•E’OiP’bS NACL.’ aC.rao GAL TREATED WA ZR
• PER çE!•IT flEATEL IN 2LENC’ •
• P siiDS nMCL/IOC’C’ GAL. 9LEt IDED WATER Q So Mt /L
S.
—4
a
—
a.1952O7193
• — 2 i.O121S9
RE IN 1•. fl776Z4
a • 64?7â11
• 4.COlflft.
a &.3519547 :.
=i.eca s3at.
• —7.29?Sà S
— 2.CO27 S328
—
N03 1. fló2 ø74â
FIGURE 53. WATER-AND COLUMN CHEMISTRY.
• • OCTOBER 19fl(loo%)
• • •, • 112:
-------
Nt•TR’.TL: N 3/L
-_L_1 I I
CHLDR o -: ME’/L
=
N I TF TE—TCJ--CHt_OR IDE EQ OO!’ ET
NOT. 1G/L IN TRE ;TED W TES
M1LE : r ‘ oo I N r Er Tao -TFF:
IDLE FEAOT ION NOT D • F E N ET R 1
(N) OF FE ENEH:4TED FEE IN
FLND Nc OL TO F:E TY .F: FF: ?M
NODE FF:4cT. ON FEE: 1 N FN
FOUNDII NN OL/OU FT TO
FOUNTS N CL.’CU FT EEE: N NEEDED
c:u. os ODL! _ :‘: . ELEE EE
=4
i9.2
= ‘ .± :
FtGURE 54. WATER AND COLUMN CHEMISTRY
NOVEMBER 1984 (100%)
FT
113
-------
SU ARY CO °AP1SOi OF CTW L CNC 1CAL
S I fl —n’rrr
U .%Itt .LLIrI rI\LLJLLU.
; tr te
Month Leik o
6-84 13
7-84 14
8-84 10
9-84 3
10-84 13
11-84 17
Averages 11.7
*
Ill ‘r n r
rLt%i I L,l1i
Salt
Lbs/Curt
5.9.1
6.36
6.46
6.48
6.35
6.53
6.35
ESTIN\TED FO 260 BY
Ni trdte Salt
Leah iqe Lbs/CuFt
17.5 5.94
15.0 6.36
12.0 6.46
11.0 6.48
10.7 6.35
1 .6 6.55
12.3 6.36
flUE
98
9.4
10.8
10.9
10.0
8.8
10.0
* From Table 23
** Froii Figures 43 through 48
+ Concentration of nitrate listed as ag N03/L.
To convert to rag fl03-N/L divide by 4.43
[ liii
8.3
9.2
11.8
II
11.3
9.7
10.3
114
-------
TABLE 25
ESTIMATES OF BUF FO FULL
BED USE i UD PARTIAL BED USE
‘ Brine Savincrs
BUF Salt Dose /Ft 3 Potential for
rionth 260 By I0 J Use 260 BV 1OO Use 1OO Bed Use
6-34 9.8 8.9 5.94 5.94 9.2
7-04 9.4 8.1 6.36 6.36 13.8
8-84 10.8 8.9 6.46 6.46 17.4
9-84 10.9 8.5 6.48 6.48 22.0
10-34 10.0 7.3 6 35 6.33 27.0
11-84 8.3 7.4 6.55 5.55 15.9
Average 17.6
115
-------
found to be at the 52% mark into the bed due to upward
displacement of bed by internals and bed geometry. This was
established by running breakthrough curves at this port and at
the bottom of the oed with fully regenerated resin and comparing
the two breakthrough points. This can be done without
discharging over 10 mg N03—N/L or 44.3 mg N03/L into the system
by stopping the flow at about 9 mg N03—N/L or 40 mg N03/L.
Computer estimated breakthrough curves can also be generated
using the method described in Reference 2. This method is not
described in this report volume. A final version of the program
and its use will be described in the final report (Vol. 2)
A series of computer produced breakthrough curves is given in
Figure 55 for the November, 1984 water composition and the resin
parameters shown in Figure 56. The resin composition data of
Figure 55 indicates the composition in equilibrium with the
effluent composition at the highest bed volume represented on
the ordinate. There are seveial. interesting features of a
similar set of curves covering a range of aed volumes up to 2000
which are not di5cussed in this report volume. However, it
should be pointed Out that if an over—run occurred these curves
show nitrate in the product water would exceed the influent
value by approximately 50 percent.
The general agreement between actual and computer estimates in
Figure 55 is quite good which indicates that the method may ne
useful for adjusting operation to obtain better efficiency. The
experimental data shows a sharper nitrate breakthrough than the
computed one. This may well be due to the use of only four
theoretical resin plates in the computations. (Computer
capacity must be increased to obtain more realistic or sharper
breakthroughs.)
It is noted that the bed is under utilized if only 260 bed
volumes are treated because it appears from Figure 55 that about
350 to 390 bed volumes could be treated for full utilization
(compare also to data of Figure 54). It is not possible to
predict from this analjsis the quantity of brine which would be
required for regeneration. A companion computer program is
under development to simulate the regeneration of spent beds
having an uneven distribution of ionic compositions.
It is also apparent that bicarbonate is present in the effluent
at the start of the run. This is not predicted from the
computer curves. Apparently not all bicarbonate is removed from
the resin upon regeneration.
The conclusion which can be drawn from these data is that the
beds display closely predictable effluent histories when using
116
-------
0 100 200 300 400 500
: - L :i+::: 11 :i:k:
80 ___ 9 ___
60
40 _____ _ _
200 _ —
BED VOLUMES
0 EXPERIMENTAL HCO PORT #2
EXPERIMENTAL N0 PORT
LI COMPUTER GENERATED
!.‘ T I., I r I ! r : F.E D--rL.3r
FiL h M!- F’ I:!;.tE: rL(rI 1’. f-I_ -,NT F! [ ’
IN C ’ r u i F ’ : i’dT:CJVE F 1 T
i n : c r . :‘ Lj- ; i cus i
I E t FLATti NI :
I-E iN CL —
F:E.!N ‘4 . •: l 1
F.E :J11 Iii_:: :
F:E iN H..0
FIGURE 55. COMPARISON OF PLANT DATA WITH
COMPUTER GENERATED EFFLUENT
HISTORIES.
117
-------
Fl t.F IO .MET: ‘;ç i
io or rr =
STi RT CONIC. r [
3Ti F’:T CiJF C.OF
Er(- RT C [ I ftJ. OF
s i- ru CUrL:
START CCNC.. OF
.r::T CO I C. OF
START ccr• o.r:’F
COHO.. SCUC F:LSkH r
(CO TO 10 £Th/L.)
IN U.N TO CL IN SOC/L.)
LU ri-i 10 i L JO
FIGURE 56. INPUT DATA FOR
COMPUTER CURVE OF. FIGURE 55.
1_L_ FF5.1 N
504 . ION 1
CL 1 ON . = j. 7
I OH =
HCL1C io i 1.
504 I:L.C I - I
NUT. SF5) N
118
-------
the computer program nase on nwrerical solutions to
mu]ticomponent equiliDria. 1 further use of tne data obtained
from computer generated effluent histories is to show the
distribution of ions in the spent oeds.
SCCONDARY PLA Vf PCRF0 1A CE FACTORS
DLlute B i n e u ntit
The McFarland Nitrate Plant nas provisions to mix saturated
brine with system water. This is accompliXdna pg ycaping
saturated brine from the brine tank, boosting brine pressure to
that of the system pressure and allowing brine tO mix with water
from the system. This mixture then flows to the brine
distributor of the vessel in regeneration. Power to the Orine
pump is controlled by the program. It the anne pump is off
fresh water can still flow througn the brine delivery system for
flushing purposes if so programmed.
The dilution of brine is desireable to prevent a slug of
saturated waste brine from entering the municipal waste wate
syste’s and to prevent osmotic shock to the resin.
The quantities of both satucated and diluted brine are given in
Table 26 together with other luantities used for eacn of the s x
months.
During August and november dilute brine quantities were reduced
to test the brine delivery system. Increased amounts of wash
water or backwash water were required to provide sufficient
washing to compensate for use of more diluted brine.
The amount of saturated brine is noted to be approximately 0.39%
of the total water produced or 0.12% of the amount treated. The
corresponding percentages for diluted urine are 0.49% and 0.65%,
respectively.
Rinses Backwash, and Total Wastewater
The amounts usea for rinsing, backwash and total wastewater are
also given in Table 26.
The plant automatically monitors conductivity of rinse water as
shown in Figure 28. Data from one such regeneration is given in
Figure 57. It appears that about 30% exceSs rinse water is used
because conductance falls to a constant value at about 2000
gallons. Rinse water was not reduced during this period to
allow sufficient rinsing. -
119
-------
T; 3LL 26
SEcO: DA Y Pt T PE F0 ”\ CE -ACT0 S
Thousand Ga1l .jris
Prcdnccd at e’ Gal 1 (mc
‘ aL. (lilute k ri e Total
Month Blond Tr ted [ sri no brine eater Bac uasri Waste
6-84 5307 3516 40 4 37670 80990 52422 171082
7-84 3595 2573 3 1 29?’O 68610 44780 142590
8 - 4 3002 2617 3772 7710 62343 47640 117600
9-8 4 5 3433 4307 22300 773a 0 57350 15690
47:3 30:5 3752 16520 67360 46133 130110
11- 31 3771 3153 012 7190 17050 31440 l15 90
Totals 2 59 i :s 226e 120760 433730 279762 83 :252
° of B1e d 103 75.0 0.00 0.49 1 76 1.14 3.30
Z of Treetc 1 133 1CO 0.12 0 6D 2.35 1.52 4 52
120
-------
WASTE BRINE
IS 30 45
MINUTES
WASTE BRINE
2500 __!__
‘ 50 °rT \
0 .L L ._
0 500 1000 1500 2000 2 00 3000 3500
GALL 0 ?1 S
FIGURE 57. WASTE WATER CONDUCT V3TY
HISTORY.
121
>.
I-
>
1500
0
1000
60
-------
The amount of backwash water is determined by the need to
declassify the resin after the brining step. If no
declassification is repiiied (when nitrate selective resins are
used), little if any mackwash would be required because of the
lack of suspended solids in the source water.
The total waste watnr over the six month period is 3.39% of the
blended water and 4.52% of the treated water. Most of this
water is not included in the amount for tne distributed water
because the rinse and backwash supply are taicen upstream of the
main meter to the distrioution system. The amounts for diluted
brine is taken from the downstieaiu of the main meter.
Therefore, these amounts must be subtracted from the blend to
give true water recovery values. Total blended water is 24598—
l2l 24477 thousand gallons pumped from the well for delivery and
an additional 834252 gallons being pumped from the source (or
waste.
Total water recovery is 96.7% over the six month period.
This hign water recovery which i even subject to improvement is
one of the main advantages of the ion exchange process over the
reverse osmosis process for nitrate removal.
on
Daily records of power consu1aptlon at the plant have been
maintained to obtain electrical power costs for well operation
and well plus mitrate plant operation. For the period starting
Januarj Li to May 23, L9 4, a total of 14750 KwF{r of power were
consumed at 4e11 2 and 6,051,000 gallons of tow nitrate water
were delivered to the distribution system. This prov .des an
avarage power requirement of 2.44 KwHr per 1000 gallons or water
delivei ed.
This total power is required to lift water from approximatelj
250 feet. below the surface . nd pressurize it to between 70 and
75 psi for pumping through the ion exchangers and delivery to a
system at between 60 and 65 psi.. The pressure drop through the
ion exchange system is approximately 10 psi. The amount of
power required to pomp water through the ion exchange vessels
was measured directly from the power meter at the higher
pressure limit. Power readings were taken with the well pumping
directly into the system and compared to readings taken wita the
plant in operation. Of the total power consumed at the site 10%
was required for the operation of the plant giving 0.244 Kwnr
per 1000 gallons the power requirement for plant operation.
Power for tne brine pump and air compressor are considered
negligible.
122
-------
The cost of this power obtained from the billing of Pacific Gas
and Electric Co. is .08183 doLlars per Kwhr making power cost
for plant operation .019967 dollars per 1000 gallons or 19.97
dollars per million gallons of blended water delivered to the
system.
123
-------
SECT iou 6
COST ANALYSES
Costs for construction of the McFarland PLant have been
previously published. (Reference 3 and 4). Operation and
maintenance costs were estimated in those references without the
benefit of Long term operation experience. The six months of
experience has indicated that only minor changes in the 06)1
costs need be made.
The capital costs are summarized in Table 27. Costs are given
for two different vessel heights. The six foot height
accommodates the three foot depth bed and the ten foot height
accommodates a five foot depth oed. The cost of the extra side
height is the most economical way of increasing bed capacity.
By increasing height from six to ten feet bed capacity is
proportiona.ely increaued. Because the diameter of the vessels
does not increase the total plant treatment capacity does not
increase. The only operational parameter which is changed is
the total gallonage which can be treated per each service cycle.
In the long term this will save wear on the valves, brine pump,
and air conpressor and the nitrate analyzer and consequently
lower the maintenance costs.
If present well water quality is maintained over what it has
been for the six month period it appears that the three foot bed
depth is adequate provided the added maintenance costs do not
become excessive.
Table 28 gives a breakdown of equipment and construction items.
These costs are taken directly from the construction contracts.
The Dionex analyzer and the control enclosure may not be
required for a standard plant installation, If the plant is
installed in cold climates, additioriul costs for enclosing the
entire plant would be required.
Revised O&M costs are given in Table 29 which reflects actual
salt and power costs for the cix month period. The costs
presented for normal plant maintenance and the miscellaneous
costs are left as previously reported as well as the resin
replacement costs. These costs may be changed if firm data on
resin loss can be obtained. No loss of resin capacity has been
detected from the operating data obtained thus far. The one
hour per day operator cost is still believed to be adequate
since this is mainly a record Keeping and inspection effort.
124
-------
TABLI 27
C\PIT L COSTS, flcFARLA O ( i s:)
It
6’D 6’9
6 ’D 1C’
I.X. Vec 1 (3 1 ’c1L cd)
96,511
111,7’.1
OC—site cor ruc on
81,151
81,15
Brire
18,700
13,700
C: ’er
40,0.5
4O ,0 5
Res! 225 Cu. ft. (3 f . p h)
35,000
624 CU. ft. (5 it. d p: .)
5’,5 C
Sub tot
$2i1,407
$309,250
3 !i r.i1C 1 15
40,711
46,385
Tc:. 1
S311,113
$355,638
* “-c a:1atu. ‘s pLan:
125
-------
T BLE 28
COST 0ET \JL FO EQuP::: T A:m C0::ST c T ::
3 Ion t - c vc sc1s 6’D ‘: 6 ’ i
÷ 4 ft. ad cd Ii
+ ‘ rc I d !cr
— SL :ot.-! ci e c--’-.
Se .cr, catch basin
E1cccrxc a1 and ca p1c card t: ’
Co.icrcte par’s
:c c:c: p r
Br:’e i np
Air c crasra:
St e i ;‘. :‘i
F1o , cc
PVC p’
Abc:c gr. c1 c:r:c’! -
Sct I.X. unites,
Set b:a-a ta-k
cccll.’ eot
o’— ,:t’ cc- :r :
Dei:er’, —-
S a total ri-e
9 ,511
8,375
6, S55
$ 6,156
2, JCG
6, ‘.00
3,6
1,227
2,373
13,900
15,795
4,667
20,603
1 ,443
863
2.200
S 61.1St .
$ 17,527
1.173
iS, 700
Czl’cr
Co.-tral parcl
$
7,614
DIcne<
12,455
Control enclesure
9,S33
Pro?r’-ab1 conrrO1jpt
Sun total other
S
126
-------
Changes in the above operating costs can be expected to drop due to
less salt used to bring the blended water nitrate level to about 30
mg/las nitrate and if resin replacement isnot required as presently
estimated. Lower salt demand can also drop due to full bed
utilization and brine reuse and/or salt recovery procedures.
Table 30 summarizes the total costs to date includin’j the above
revisions. The amortized annual capital cost per 1000 gal. is based
on 100% use of the l—mgd capacity. The McFarland plant was only
operated at 13.7% of its fulicapacityduring this initialperiod (See
Table 20). In this case the amortized annual capital cost par 1000
gal is 7.30 times that shown in Table 30 or $.832 per 1000 gal. F s
annual plant production falls from 100% to 0% of full capacity use this
cost rises from $0.1l4 to infinity. 0&M costs per 1000 gal are
estimated to remain approximately as given in Table 29 regardless of
plant useage. The high cost of capital amortization of a partially
used plant must be taken into consideration when assessing the cost
impact on the consumer. The true water cost which the consumer must
pay for operating the plant at less than full capacity can be estimated
by comparing consumer costs with and without the plant.
Trueconsumerwatercost sfor thisreport periodreflect the factthat
the consumer receiveswater frontheplant aswellas fromother wells
in the system. In this case the capital costs associated with water
supply captial costs in McFarland is the capital cost of wells, the
distribution system, and related facilities and improvements (not
including a nitrate plant), CS, plus the capital costs shown in Table
30,CP. Thetotaleonsu lvercostofamOrtizingthecapitalCOsts (per
1300 gal of water consumedj by producing a fraction of l—mgd from
existing facilities and the remainder from the nitrate plant is:
Total capital cost/1,000 gal = (CDS ÷ CP)/ 1,000 gal.
The additional annual amortized capital cost which the consumer must
pay for partial (or full) use of the nitrate plant is the amortized
capital cost, $41,773, for the nitrate system as shown in Table 30.
The added cost due to O&M of the nitrate plant during this report
period is 0.137 tines the O&N cost of Table 29.
The total added consumer cost durirg this report period due to nitrate
treatment of 13.7% of the water supplied to the system is:
$/l000 gal = $0.l14 + 0.137 x $0.l3l
or $0.162
These cost analyses will be presented in more detail when all costs
over a two yeaz pariod of operation are available and will be discussed
in Volume II of this report.
127
-------
The impact of the plant’s operation on the cost of water to the
coamunity can be assessed by comparing the rates given in Table 31
which were established prior to the ConStruction of the nitrate
plant. Itisestimatodthattheaverageosageofwater inMcFarland
over a one year period is 15,000 gallons per mont’i per connection.
This gives the average rate of $.731 per 1000 gallons paid by the
consumers. Water costs given in the above tables do not include the
normal pumping costs which are estimated to be about $.1B per 1000
gallons.
Total cost of water from Well 2 including electrical power to pu ’p and
pressurize water from below the surface plus that required tooperate
and maintain the plant (full time) would be $.425 per 1000 gallons.
This is well within the S.711 average rate charged to the users and
allows $.306 per 1000 gallons or 41.8% for administration and
nsge’ t of the water system.
128
-------
TABLE 29
OPERATION AND MAINTENANCE COSTS
Item Annual SJ lOOp , , ,gai
Operation ( 1 hr/day) 4,745 0.013
Normal O&M, .02X(355,628) 7,133 0 .019
*po ,Jer, boost purnp (.093.kwh) 7,289 0.020
Resin replacement (5 yrs) 11,522 0.032
**SaIt ($31.50/ton), 7O cap. 14,314 0.034
Miscellaneous 3,000 0.008
TOTAL OMI 47,983 0.131
TABLE 30
TOTAL COST OF PLANT AND COMPARISON
TO PRESENT WATER COST IN McFA( LAN0
Total Costs Cap ital+0 ,)
Annual Co ! (S) $/ QQPLa i.
Capital costs - 5355,638 (20 yrs. 10’) 41,773 0.114
Operet on & mathtenance 0.131
89,756 0.245
TABLE 31
PRESENT COSTS OF WATER IN McFARLAND
Sf1000 gals.
First 11,000 gals. 0.980
Over 11,000 gals.
* 244 kWh per million gallons based on .O8183 per kwh.
** 2490 lbs. per million gallons.
129
-------
REFERENCES
1. Guter, G.A. ‘Removal of Nitrate from Contaminated Water
Supplies for PubLic Use, Final Report,” Murucipal
EnvironmentaL Research Laboratory, Office of Research and
Development, U.S. EPA, Cincinnati, OhLo. EPA—600/2—82-
042, August 1982.
2. Guter, G.A. NEstimatlon of Resin and Water Composition On
Coiumn Performance In Nitrate Ion ixchange ”, f WWA 1984
Annual Proceedings, Dallas, Texas, June 10—14, 1984, pages
1631 to 1649.
3. Lauch, R.P. and Guter, G.A. “A One ‘lGD Ion xchange Plant
for Rernoval of Nitrate from Well Water”, ibid, pages 713
t 733.
4. Guter, G.A.
McFarland,
Proceedln9s,
Nevtda, June
“Operation Performance and Cost of the
CA Nitrate Removal Plant”, AWWA Sei inar
Control of Inorganic Contaminants, Las Vegas,
5, 1983, Pages 29—51.
130
-------
------- |