200501051st Response of United Water No 9.pdf

UNITED WATER IDAHO INC.
CASE UWI-O4-
FIRST PRODUCTION REQUEST OF THE COMMISSION STAFF
Preparer/Sponsoring Witness: Scott Rhead
Telephone: [phone]
Title: Managing Engineer
REQUEST NO.
Reference p. 12 , lines 14-23 and page 13, lines 1-3. Please provide a copy of
the 1/8/02 Basis of Design Report.
RESPONSE NO.
See attached document:
Basis of Design Report.
UNITED WATER IDAHO INC.
CASE NO. UWI-W-O4-04
FIRST PRODUCTION REQUEST
IPUC STAFF
ATTACHMENT TO
RESONSE TO
REQUEST NO.
Columbia Water
Treatment Plant
MWH
. ..
MOIIrGOMOI'WAm.\'-
MWH
MQ,"fliTGOMf,!=i!'t"'f'llA150N HljRZA
January 8, 2002
Mr. Greg Wyatt
United Water Idaho
8248 West Victory Road
Boise, ID 83707-1488
Subject:Columbia Water Treatment Plant
Basis of Design Report
Dear Mr. Wyatt:
We are pleased to provide eight (8) copies of the Basis of Design Report for the Columbia Water Treatment Plant.
This document provides the supporting information for our recommendation to United Water Idaho for the
selection of the dissolved air floatation (DAF) followed by Ultra Membrane Filtration (UF) for the proposed
Columbia Water Treatment Plant. The Basis of Design Report contains the following:
Executive Summary
Workshop No.- Water Quality Technical Memorandum
Workshop No.- Treatment Technology EvaluationTechnical Memorandum
Workshop No.- Process Section Technical Memorandum
Pilot Study Report - The Effect ofDAF Pretreatment on Membrane Performance
The Executive Summary provides the recommendation for the new Columbia WTPfacilities and a summary of the
pilot study report, each workshop Technical Memorandum and the 10% Conceptual Design Plan for the Columbia
WfP. Incorporated into each Technical Memorandum are all water quality data and figures, regulations, project
objectives, design criteria, and comments from the United Water Idaho staff based upon their participation and
review of the project deliverables.
If you have any questions or concerns, please contact us.
Respectfully submitted
Edwin T. Cryer
Vice President
Ips
671 Eust Riverpurk Lune
Suite 200
Boise, Idaho
83706-4000
Tel: [phone]
Fox: 208 345-5897
Delivering Innovative Pro;ects and Solutions Worldwide
CO L UMBIA WATER TREATMENT PLANT
BASIS OF DESIGN REPORT
TABLE OF CONTENTS
Executive Summary
Technical Memorandum No.
Technical Memorandum No.
Technical Memorandum No.
The Effect of DAF Pretreatment on Membrane Performance
BASIS OF DESIGN REPORT
EXECUTIVE SUMMARY
BACKGROUND
Montgomery Watson Harza (MWH) in association with Carollo Engineers conducted three
specific technical workshops to provide the basis for process and technology selection and
establishment of design guidelines proposal for the United Water Idaho (UWI) Columbia Water
Treatment Plant. Three Technical Memoranda were written to present the results of each
workshop. These Technical Memoranda are as follows:
Workshop No.1- Water Quality
Workshop No.2 - Treatment Technology Evaluation
Workshop No.3 - Process Selection
The workshops and associated Technical Memoranda provided the basis for selecting dissolved
air floating (DAF) and membrane treatment as the preferred process alternative. A pilot study
was conducted at the Marden Water Treatment Plant (MWTP) to further evaluate this selection
and determine the influence of DAF pretreatment on UF membrane performance. The results of
the pilot study are included in The Effect of DAF Pretreatment on Membrane Performance report
prepared by MHWI Carollo. Water quality samples were taken at the Boise River intakes located
at Marden WTP and Highway 21 bridge pump station throughout the pilot testing period for
comparison of water quality characteristics at the two locations. The actual Columbia WTP
Boise River water intake will be the Highway 21 pump station. The results of the water quality
analysis are presented in the pilot study report as Appendix A, Tables A-I and A-2. The
Leopold DAF Pilot Plant Report is located in Appendix B of the pilot report. A summary of
each of these Technical Memoranda and the pilot study report is provided at the end of the
executive summary.
These workshops and subsequent Technical Memoranda; the results of the pilot study and water
quality analysis; plus a number of meetings with UWI staff resulted in final selection of the UF
membrane filtration with DAF pretreatment followed by ultraviolet light (UV) disinfection
option. The design criteria and process flow diagram included in Workshop No.3 - Process
Selection Technical Memorandum, defined the Basis of Design of the Columbia Water
Treatment Plant. A conceptual site and facility layout of the Columbia WTP is shown on Figure
ES-
The recommendation for delivery of Boise River water to the Columbia WTP site includes
improvements to the Highway 21 Bridge Intake facility to provide the initial 6 mgd plant
capacity, expandable to 20 mgd. These improvements include the addition of pumps, associated
mechanical and electrical appurtenances and controls, and segments of the 30-inch diameter raw
water transmission pipeline to complete the delivery conduit. ANew York Canal diversion, flow
control and system metering to the intake wet well is also required.
ES-
FI
G
U
R
E

E
S
-
1
10
1
.

C
O
N
C
E
P
T
U
A
L
DE
S
I
G
N

P
L
A
N
SC
A
l
E

IN

F
E
E
T
WA
T
E
R
PO
T
A
B
L
E

W
A
T
E
R

P
I
P
E
L
I
N
E
3/
4
8
PO
T
A
B
L
E

W
A
T
E
R

P
I
P
E
L
I
N
E
3/
4
6
3/
4
4
WORKSHOP PROCESS SUMMARY
Workshop No.1 - Water Quality
Introduction. Technical Memorandum No.1 presents the results of a water quality workshop
conducted in Boise, Idaho offices of Carollo Engineers on January 31 2001. The workshop goal
was to evaluate available Boise River raw water quality information and the finished water
objectives for the Columbia WTP.
Boise River Raw Water Quality. A summary of Boise River water quality was developed
based upon data collected from the new river intake for the Columbia WTP and the existing
Marden WTP. Specific summaries are then given for each type of measurement performed
through figures, tables and text. General background information and site-specific information
are given for turbidity, total organic carbon, temperature, pH, microbial ICR data, disinfection
by-products, tastes and odors, and algae measurements.
Finished Water Quality Goals. Water quality objectives for the Columbia WTP are presented
in tabular format. These water quality goals included primary standards for turbidity, Giardia
virus, and Cryptosporidium removal and maximum values for THMs, HAAs, bromate, and
chlorine dioxide. Goals for secondary standards were included for taste and odor, color, iron and
manganese.
Summary. The summary states that the Boise River is a high quality source for raw water
supply, which allows selection of an efficient and cost-effective treatment process.
Workshop No.2 - Treatment Technology Evaluation
Introduction. Technical Memorandum No.2 presents the results of a technology evaluation
workshop held in Boise, Idaho at the office of Carollo Engineers on February 1 , 2001. The
workshop objective was to screen a comprehensive list of available technologies to obtain alimited short list of two or three candidate water treatment processes, at most which are
appropriate for the proposed Columbia WTP. Appropriate processes are those capable of
treating the raw water quality to meet the finished water objectives defined in Workshop No.
(documented in Technical Memorandum No.1).
Technical Evaluation Criteria. Two filtration options are recommended for evaluation in this
memorandum - granular media filtration and membrane filtration. Each filtration method was
evaluated in terms of meeting final water quality objectives for the turbidity parameter. The
treatment alternatives were also assessed for their effectiveness in meeting tastes and odors
algae, TOC removal and disinfection of microbial contaminants (virus Giardia
Cryptosporidium) standards and objectives established in Technical Memoranda No.
Conclusions of Treatment Technology Evaluation. Table 2.1 in Technical Memoranda No.
(attached) presents a summary of treatment technologies evaluated in this workshop. The
selection of the final treatment process between the granular media and membrane filtration
ES-
options will be based upon refined design criteria and cost estimates as presented in Workshop
No.3 and documented in Technical Memorandum No.
Workshop No.Process Selection
Introduction. Technical Memorandum No.presents the results of a process selection
workshop held in Boise, Idaho at the office of Carollo Engineers on February 23 , 2001. Theworkshop objective was to select the recommended treatment process for the proposed Columbia
Water Treatment Plant (WTP). Preliminary design criteria and planning level cost estimates forthe two alternatives described in Workshop No.2 (granular media filtration and membrane
filtration) were presented and reviewed during the workshop.
Review of Water Quality and Process Screening. Water quality issues and preliminary
screening of treatment process alternatives considered during Workshops No.1 and 2 were
reviewed briefly. Additional comments and clarification from the Marden WTP operational staff
were provided for turbidity, total organic carbon, taste and odor, and algae.
The potential use of the Columbia WTP as a source of water for aquifer storage and recovery
(ASR) was discussed. Based upon the capacity requirements selected, ASR production is not a
significant criteria for selection of the treatment process options at the Columbia WTP.
Dissolved air flotation pretreatment was discussed as the best pretreatment for Boise River water
quality. The granular media filtration and membrane filter options are discussed and
summarized by text, figures, tables, and a cost estimate. Text discussions for each option includeriver intake and raw water pump station, flash mix, flocculation, DAF basins, filters, filterbackwash, disinfection, solids handling, and chemical feed.
Design criteria and process flow diagrams are presented for the grandular media and membrane
filtration options. Both options include DAF as the selected pretreatment option. The design
criteria are based on an initial flow of 6-mgd in the first phase of construction with ultimate plant
capacity of 20-mgd. The first phase of 6-mgd was selected by UWI to meet the anticipated
demand for potable water in the Columbia Bench! Gowen Service Zone. The preliminary designcriteria define the equipment characteristics for the raw water pump station, flash mix
flocculation, DAF basins, filters, filter backwash, disinfection, solids handling, and chemical
feed.
Cost Analysis. Table 3.3 (se Technical Memorandum No.3) presents the estimated capital costs
for a 6-mgd capacity Columbia WTP. Comparative costs are presented for both the granularmedia filtration option and the membrane filtration option. The costs presented have a planning
level of accuracy (+ 30%).
Alternatives Evaluation. Evaluation criteria were selected for the evaluation! comparison ofthe two proposed process options for the Columbia WTP. The six criteria with the greatestweight in process selection are presented. These six criteria are reliability (including finished
water quality, regulatory performance, reliability during unmanned operation), capital costs
operation and maintenance cost, modularity for expansion, and production flexibility.
ES-
Process Recommendation. The recommended treatment process for the Columbia WTP is the
membrane filtration option with dissolved air flotation pretreatment and UV disinfection.
THE EFFECT OF D AF PRE TREATMENT ON MEMBRANE PERFORMANCE
REPORT SUMMARY
Introduction
A pilot study was conducted to evaluate the effectiveness and establish design criteria of the
preferred treatment process option of DAF pretreatment and UF membranes. The main goal of
the pilot-study was to determine if DAF pretreatment would allow the membrane filters to be
operated at a higher flux rates in comparison to using raw Boise River water. This report details
the findings of the study.
Background
The process of membrane filtration is described in the Pilot Study report. Despite the potential
benefits of low-pressure membrane filtration, capital costs can be higher than conventional
granular media filtration followed by disinfection technologies. In order to reduce these high
costs, pre-treatment systems are often used upstream of the membrane.
Reduction of suspended and particulate matter, including total organic carbon and algae, from
the membrane feed was a main goal of the DAF pretreatment process for the Boise River pilot
study. The report provides a description of the DAF process. The experimental approach of the
study was discussed, including the membrane performance objectives such as flux, flow rate
backwash frequency and percent recovery.
Methods
Description of Pilot System. The DAF pilot assembly consisted of a two-stage flocculation
basin followed by a dissolved air flotation unit. Ferric chloride was the coagulant used in the
evaluation. For most of the study, hydrochloric or sulfuric acid were added for pH adjustment to
the DAF influent prior to the water entering the flocculation basin. No polymers or other
chemicals were used in the study as their use generally has a negative impact on DAF andmembrane performance.
Water Quality Testing. The water quality tests performed during the first phase are present in
Table 1 of the Pilot Plant report. Turbidity and pH of the DAF influent and effluent were
recorded continuously by on-line instrumentation. Raw water temperature and dissolved oxygen
were also recorded continuously.
Pilot Operational Parameters and Laboratory Analytical Results. Table A-I of Appendix A
of the Pilot Plant report presents the daily averages of the operational parameters used in the
August pilot study with DAF pretreatment. This table also presents the TOC, DOC, UV -254 and
disinfection by-products (TTHM and HAAS) analYtical results measured at specific points over
ES-
the course of the study. Results from the analYtical tests of the Dec. 26, 2001 samples and the
January 2002 samples have not yet been received from MWH Laboratory.
Results
Water Quality Characterization. Throughout the first phase of the pilot study, the ferric
chloride and acid doses were varied to optimize the DAF effluent turbidity. The results of the
water quality analyses conducted during the first phase of the pilot-study are presented in Table 2
of the report. The data provided in Table 2 for the membrane permeate water indicates that the
membrane filtration process removed little to no dissolved organic matter based on the DOC and
UV 254 absorbance analyses. Water quality analysis results from the second phase of the pilot-
study are presented in Table 3.
Hydraulic Performance of UF Membrane with DAF Pretreatment (August 2001). Thehydraulic performance of the membrane was measured continuously by on-line instrumentation
on the membrane pilot plant. The parameters measured by the membrane pilot instrumentation
were transmembrane pressure (TMP), specific flux normalized flux, and temperature. Themembrane run termination criteria established for the study was a TMP of 15 psi or a 30-daychemical cleaning interval, whichever came first.
Membrane hydraulic performance data for the first phase of the study is presented in Figure 1 of
the report.
Hydraulic Performance of UF Membrane on Raw Boise River Water (September 2001).
Following the initial phase of testing with DAF pretreatment two sets of membrane tests were
conducted using raw Boise River water. The operational data obtained during the first raw water
membrane run is provided in Figure 2 of the report. The data provided from the second run is
provided in Figure 3.
Figure 4 of the report is a graphic representation of the influence of coagulation on membrane
performance. Evidence of enhanced backtransport and reduced pore penetration can be seen in
Figure 5 (with DAF pretreatment). Over the course of the DAF/membrane testing, some
membrane fouling did occur; however, compared to the prior raw water testing without DAF
pretreatment, the fouling rate was significantly lower. A very low three-hour TMP trend (time to
target pressure change) for the first raw water condition is presented in Figure 6.
CONCLUSIONS
Based on the results of this study, it can be concluded that the DAF pretreatment significantly
increased the treatment capacity of the UF membrane filtration module used for this study.Under the conditions tested using DAFpretreatment, the membrane TMP remained below 12 psifor the 25-day run. The data suggest that possibly higher flux rates (greater than 68 L/m2 -h ~
200C) could be achieved with DAF pretreated water even at the high surface loading rate of 8
gpm/sq. ft, while maintaining the 30-day chemical cleaning interval. During the second phase ofthe pilot-study, a decrease in membrane performance using raw Boise River water without
pretreatment was clearly shown. At a flux of 68 L/m2 -h ~ 20o , terminal TMP was reached in
ES-
5 days (11.5 hours), and at a flux of 55 L/m2 -h ~ 20o , terminal TMP was reached in 1.5 days
(36 hours). See Figure 7 of the report.
ES-
TECHNICAL MEMORANDUM NO.
TECHNICAL MEMORANDUM NO.
TECHNICAL MEMORANDUM NO.
L M M N
. I
- j
(I)MONTGOMERY WATSON CarOLLO" e
"'"
UNITED WATER IDAHO - COLUMBIA WATER TREATMENT PLANT
, -
To:Greg Wyatt
Dan Brown
Reference:6232A.
. !
From:Matthew Marshall
Silas Gilbert
Date:December 21 2001
Final
Subject: Workshop No.1 - Water Quality
- ,
INTRODUCTION
. This technical memorandum presents the results of a water quality workshop held in
Boise, Idaho at the office of Carollo Engineers on January 31 , 2001. The workshop goal
was to evaluate Boise River raw water quality and finished water objectives for the
Columbia WTP.
' I
The Boise River is a high quality source of supply that requires little treatment to meet
water quality regulations - relative to water supplies in many parts of the country. The
worksh6p goal of evaluating quality and treatment objectives is essential to a wise use
of funds in development of the proposed- water treatment plant.
Attendees:
Bob Raczko
. Dan Brown
Scott Ahead
Bill Carr
Bob Adams
Joe Jacangelo
Dan Askenaizer
Si Gilbert
Bill Lynard
Gil Crozes
Bryant Bench
Matthew Marshall
Chris Cleveland
Susumu Kawamura
Monty Marchus
United Water -Resources
United Water Idaho
United Water Idaho
United Water Idaho
United Water Idaho
Montgomery Watson
Montgomery Watson
Montgomery Watson
Montgomery Watson
Carollo Engineers
- Carollo Engineers
Carollo Engineers
Carollo Engineers
Kawamura Water Engineering
Idaho DEQ
. -
. I
.. J
.. j
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
BOISE RIVER RAW WATER QUALITY
, !. ,
Raw water quality and finished water objectives determine process selection. Raw water
quality data for the Boise River were colleCted from a variety of sources, including the
new river intake for the Columbia WTP (immediately west of the Highway 21 bridge
crossing) and the existing Marden WTP. The new Columbia WTP intake and the Marden
WTP intake are located approximately 2.5 miles and 8.5 miles, respectively,
downstream of Lucky Peak Dam.
In general , the water quality in the Boise River improves in the upstream direction. The
inclusion of Marden WTP data is slightly conservative because the water quality is
generally better at the Columbia WTP river intake , which is approximately 6 miles
upstream of the Marden WTP intake. The Marden WTP raw water data were segregated
to separate the Boise River data from the Ranney collector supply data. In most
respects, the Ranney supply is of higher quality than the river supply at the Marden
WTP. The Columbia WTP will not have the benefit of a Ranney collector supply.
. ,
The water quality data presented herein include the parameters that are most important
in treatment process selection. In some cases, quantitative data are not available and
the Marden WTP operations staff furnished qualitative information. Additional water
quality data collected as part of the OAF! Membrane Pilot Study conducted August
through September 2001 are presented with the pilot study results.
Turbidity
, ;
Turbidity is a measurement of the light scattering or light absorbing properties of water.
Turbidity in drinking water supplies is commonly caused by the presence of suspended
matter, such as clays, silts, finely divided organic and inorganic matter, plankton, and
other microorganisms; A seasonal profile of Boise River turbidity (at both the Marden
and Columbia intake locations) for a period of two and one-half years is presented in
Figure 1.1. The same data are presented as a frequency distribution in Figure 1.
The Boise River is a high quality source in respect to t~rbidity. This is primarily due to
the removal of suspended solids in Lucky Peak Reservoir and the other upstream
impoundments. The 50th percentile turbidity is approximately 4 NTU and the 95th
percentile turbidity is less than 8 NTU. The Marden WTP operations staff states that the
highest Boise River raw water turbidity measured at the existing plant is 17 NTU.
, .
Total Organic Carbon
Total organic carbon (TOC) is a measure of the organic carbon , both particulate and
dissolved , in a water supply. TOC is a useful parameter in gauging the general level of
organic constituents in a water supply. Some TOC constituents are precursors to the
formation of regulated disinfection by-products. TOC can contribute to fouling of a micro-
or ultrafiltration membrane.
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
, I
. J
- ,~ .. !
: 1
. ,. ~- -
- J
. j, j
. J
, J
. j
. .i
- );::....
"'C
:c 12100.
:::J
98 6-98 9-98 12-98 3-99 6-99 9-99 12-99 3-00 6-
Date
BOISE RIVER TURBIDITY
Figure 1.
" ", .._
, J
. j
- 1
" \. ..
, T 1
- j
- i
, i
. I
. - j
,.. J
. 1
, J
. J
, J
. !. j
100%
(I,) 60%
(I,)
+oil
. z
. +oil
(I,)
40%
80%
20%
10 12 14
Turbidity (NTU)
BOISE RIVER TURBIDITY
Figure 1.
c - ,
C' -
. -- j. j
. 1
, j. !
. J
! .
; J
. j.- j
A seasonal profile of Boise River total organic carbon (TOC) data at the Columbia wrp
intake for a period of approximately four years is presented in Figure 1.3. The same
data are presented as a frequency distribution in Figure 1.4. Additional data collected in
the spring and early summer of 1990 as part of the Marden wrp pilot plant studies are
, presented in Figure 1.
Except for short-term, seasonal episodes, the Boise River is a high quality source with
respect to total organic carbon. Episodes of elevated TOC appear to be associated with
the flushing of the river by high flows at the beginning of the irrigation season. Typically,
. high TOC events are less than a week in duration. The 50th percentile TOC is 1;7 mg/L;
the 95th percentile is less than 2.5 mg/L.
Temperature
Some water treatment processes are affected by water temperature. Physical processes
are affected because water viscosity and density increase as temperature decreases.
Chemical processes are affected because solubilities and reaction kinetics change with
temperature. Biological processes, including disinfection , are also affected by
temperature.
A seasonal profile of Boise River temperature for the period of 1998 through 2001 is
presented in Figure 1.6. The same data are presentec;l as a frequency distribution in
Figure 1.7. These data were collected at the intake for the Columbia wrP.
The maximum reported temperature is approximately 68OF; the minimum temperature is
approximately 32OF. The 50th percentile temperature is 46OF. The seasonal variation in
temperature is important because the specific flux of a membrane filtration process
capacity is reduced by temperature; therefore, the capacity of a membrane filtration
plant can vary seasonally with water temperature.
pH is an expression of the hydrogen ion concentration in water. A pH of 7 represents a
neutral condition; a pH greater than 7 represents a basic (alkaline) condition , and a pH
less than 7 represents an acidic condition. pH is an important parameter governing
many chemical reactions in water treatment including disinfection and the formation of
regulated disinfection by-products.The alkalini~y (or buffering capacity) of a water
supply moderates changes in pH.
A seasonal profile of Boise River pH, collected at the Columbia wrp inlet, for the period
of March 1998 through July 2000 is presented in Figure 1.8. The same data are
presented as a frequency distribution in Figure 1.
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
- I
c 1
- j
. I
. ~
. 1
c. j
....J
s:::
..c
s:::
....
0/28/95 3/11/97 7/24/98
BOISE RIVER Toe DATA AT PUMP STATION
Figure 1.
12/6/99 4/19/01
, J
. "I
- ,
. 1
i -
- J
,-,
, 1
100%
s:::
60%
s:::
40010
c...
20010
- 0%
80%
TOC (mg/L)
BOISE RIVER TOC
Figure 1.
C";
. ,. "
, f
..
...J
-...(!)
::E - 6
(.) .. )
. J
Day = 0 1 0 - 2 0
..-
27MAR
.l&Qend-0- TOC--0-. UV Absorbance
1 MAY
60
1 JUN
BOISE RIVER TOC AND UV ABSORBANCE
1990 PILOT STUDY OAT A
Figure 1.
90 100
1 JUL
o. 13
12 v
L()
C\I11
1 0
09 c::(
08 a
07 c::(
06
. ,., \
, V
- J
. )
!::. 40
........
c..
. I-
. !
0 -1.8
98 6-98 9-98 12-98 3-99 6-99 9-. 12-99 3-00 6-
Date
BOISE RIVER TEMPERATURE AT PUMP STATION
Figure 1.6
4 ~
........
c..
CD.
. \
. ;1
, J
. j
. t
. I
- )
. i
100%
=c 60%.(I)(I)
(,)
15 50%
40%
(I)
90%
80%
70%
30%
20%
00/0
Temperature (F)
BOISE RIVER TEMPERATURE
(1998.2000 DATA)
Figure 1.
. J
, !. '\
. J :t::
::s 7.
...
1n 7.
. .
:J:c..
- J
". f
. j" )
6.4
98 - 6-98 9-98 12-98 3-99 6-99 9-99 12;.99 3-00 6-
Date
BOISE RIVER PH AT PUMPHOUSE
Figure 1.
0%
6 6.7 6.8 6.9 7.0 7.1 7.3 7.4 7.5 7.6 7.7 7.8 7.9 8.
100%
. ,
90%
80%
. 1
70%
. ,
s:::
:c 600
C1)
C1)
Co)
50%
...
C1)
40%C1)
D..
. ,
30%
20%
. J 10%
- J
BOISE RIVER pH
(1998- 2000 Data)
Figure 1.
o. j
- )
- J
. j. j - ., )- \. .
The 50th percentile pH is 7.2: the 95th percentile is 7.5. The Marden WTP operations
staff report that the water is poorly buffered, with an alkalinity of less than 30 mg/L as
CaCO3 in the spring and approximately 40 mg/L as CaCO3 in the fall. The plant staff has
observed a diurnal pH variation during the summer. Reportedly, the pH swings from 6.
. to 7.3 with the higher pH coinciding with maximum water temperature and sunshine
(mid-afternoon). It is likely that the daily swing in pH is due to the uptake and release of
carbon dioxide by aquatic vegetation and algae.
MicrobiallCR Data
Microbial contaminant data collected from July 1997 through December 1998 as part of
the Information Collection Rule (ICR) include the following:
A monthly profile of Giardia cyst sample results is presented in Figure 1.10.
Monthly "results were generally low, either no cysts detected or less than 10 cysts
per 100 liters; the highest single sampling event result was approximately 275
cysts per 100 liters.
Cryptosporidium oocysts were not detected in any monthly sampling during ICR
data collection.
A seasonal profile of virus sample results is presented in Figure 1.11.The monthly
sampling results fluctuate between no virus detected and 1 MPN per 100 liters.
A profile of monthly total coliform sample results is presented in Figure 1.12.
Monthly samples ranged from none detected to over 1 000 total coliforms per 100 mL. .
, .
A profile of monthly E. coli bacteria sample results is presented in Figure 1.13.
The absence of fecal coliform during high total coliform events may be consistent
with the absence of Cryptosporidium oocysts.
There is insufficient data tq est?,blish causes for variations in microbiaLwater quality-
parameters. The purpose for presentfng the data is to establish a range of water quality
to be treated at the Columbia WTP rather than to propose watershed control or
protection practices.
. Disinfection By-Products
Seasonal profiles of THM and HAAS sample results collected for the ICR are presented
in Figure 1.14. These data represent samples from the distribution system supplied by
the Marden WTP. Several factors diminish the usefulness of these data as a direct
comparison to the proposed Columbia WTP:
The Marden WTP data include both Boise River surface water and Ranney
Collector water as a blended source of supply; the Columbia WTP supply is theBoise River only.
The Marden WTP distribution system service area is a blended supply with a
number of groundwater wells contributing. The groundwater available for the
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
,. ., .. ,
I '300
250
. i
200
, ". j
en 150
. !, )
100
3~97'
.. )
BOISE RIVER GIARDIA CONCENTRATIONS
Figure 1.
11-
. j". ,, -.... !
- i ...J
~ '
0.4
. ~
11-
BOISE RIVER VIRUS CONCENTRATIONS
Figure 1.
, .. ). .- ;. ,. j. )
- I
. j. j
10000.
1000.
..J
L0-
100.
10.
11-
BOISE RIVER COLIFORM CONCENTRATIONS
Figure 1.12
- J
::J 15
:z
ui 10
- j- .. "
11-
BOISE RIVER E. COLI CONCENTRATION
Figure 1.
.. ,. j! .
-.- THMs (ppb)
--- HAA5 (ppb) ,
c .
DBP data taken from Marden WTP treating Boise River water and Hann~y
water in combined distribution system with groundwater wells.
~ j
- .I DBP DATA
ICR MARDEN WTp
Figure 1.
11-
" I
service area of the Columbia WTP is more limited and treated Boise River surfacewaterwill predominate.
The Marden WTP is able to switch to Ranney Collector water, or limit production
during periods of poor Boise River water quality; the Columbia WTP will be less
flexible in it~ ability to reduce production and will depend solely on the Boise River
as a supply.
" f
" I
Formation potential for THMs in raw and treated Boise River water collected in 1990
during the Marden WTP pilot study are presented in Figure 1.15. The THM formation
potential presented in this figure is based on an eight day holding time with. excess
chlorine (following General Water Works Company protocol at the time). United Water
Idaho staff calculate that the overall detention time in the Columbia WTP service area
(Columbia bench) is less than 24 hours.
From Figure 1.15 it can be seen that the raw water THM formation for 24 hours is less
than 80 ~g/L; the treated water THM formation (at a 25 mg/L ferric chloride coagulant
dose) is less than 20 ~g/L. Based on the following assumptions it is proposed that OBP
formation will not likely be a problem at the proposed Columbia WTP even with the
continued use of free chlorine as a distribution disinfectant:
" i
Limited detention time in the Columbia bench service area.
The potential for some OBP precursor removal through the treatment process
(even -direct filtration or membrane filtration).
- j
THMs as the critical OPB (rather than HAA5, etc).
No significant deviation in water quality from previous sample data.
. .
Additional OBP data were collected during the OAF/Membrane Pilot Study (August
September 2001), as shown in Ta~le 1.1. The -samples were dosed with chlorine and
held under a simulated distribution system (SOS) protocol.
- J
G :\U n ited Water\IDAHO\6232a .OO\Correspondence\EM-OO 1 . doc
J,.eoenrt--a- Raw Water
Fift8red Water (Ferric Chloride. 25 mgll,
Polymer . O~ 15 mgJ1)
. j
200
180
160
140
120
100
. )
100
I.L
a..
C/J
J.J
...
::J
:i!
....-----------
.JII.
...
"'I""
I"'"'-;r .....I
---. .
. 1
............"",.. --'".... .-
1000"
,..,."..'--" "" .,.......
r::
. 1 . 1 10
TIME (DAYS)
BOISE RIVER RAW AND TREATED WATER THM FORMATION CURVES
1990 PilOT STUDY DATA
Figure 1.15
, .
. i
, !
~ 1
" J
. I
~ ", !
Table 1.Boise River DBPs
Columbia Water Treatment Plant
United Water Idaho
Sample Date Sample SDSTHMa
(Jlg/L)
SDSHAASa
(Jlg/L)
8/18/01 Membrane permeate with OAF
p retre atm e ntb
Membrane permeate with OAF
pretreatmentC
Membrane permeate without
pretreatment
SOS Protocol: 48 hours chlorinated holding time; 2 mg/L chlorine. dose, Final
chlorine residual approximately 1 mg/L prior to quenching after holding time; final
pH unknown.
Pretreatment: OAF flocculation time = 20 minutes; flocculation GT = 60 000 sec
OAF surface loading rate = 4 gpm/sq ft; ferric chloride dose = 8 mg/L; sulfuric acid
dose = 19 mg/L.
Pretreatment: OAF flocculation time = 10 minutes; flocculation GT = 30 000 sec
OAF surface loading rate = 8 gpm/sq ft; ferric chloride dose = 10 mg/L; sulfuric
acid dose = 16 mg/L.
8/29//01
9/25/01
The data show that Boise River water, even without pretreatment (i.e., without
coagulation), does not exceed anticipated THM or HAA standards under SOS
, conditions. However, HAA levels are proportionally higher than indicated by previous
ICR data for the Marden WTP. Additional SOSDBP sampling of Boise River water is
recommended.
Tastes and Odors
There is limited quantitative data available for tastes and odors in the Boise River
supply. However, Marden WTP operational staff reports several instances of tastes and
odors occurring in the spring or in the fall resulting in some customer complaints. The
duration of these events is typically two or three weeks at the longest. Based upon
information furnished by United Water Idaho staff, the presumptive taste and odor
compounds are geosmin and MIB.
The method of treatment for taste and odors at the Marden WTP is powdered activated
carbon (PAC). Doses have been limited to less than 10 mg/L in order to avoid excess
filter breakthrough of PAC when the Marden WTP is operated in a direct filtration mode.
Typically, the Marden WTP pretreatment (Superpulsators) has not been operated during
taste and odor episodes. Taste and odor control is an important issue for process
selection.
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
Algae
Limited quantitative algae data are available for the Boise River. The results of a sample
taken during the 1990 Marden WTP pilot program are presented in the Table 1.2. United
Water Idaho staff collected additional data during the summer of 2001 , as shown in
Figure 1.16.
- J
Table 1.Boise River DBPs
Columbia Water Treatment Plant
United Water Idaho
Sample Date Organisms/ml Species
4/12/90 900 Predominant
A sterian ella
Others
Diatoms (esp. Fragillaria)
Stephanadiscus, Flagellates, Ulathrix
Chiarella
. ,
Qualitative evidence presented by the Marden WTP staff indicates that algae can be a
moderate to serious problem on occasion. On at least one occasion , the presence of a
filamentous alga was significant enough to require flushing the 3/8-inch mesh
mechanical screens every several hours. During this event, filter production was limited
by the head loss created by the filter clogging algae. Process flexibility for handling algae
blooms is considered an important issue for process selection at the Columbia WTP.
The Columbia WTPwili not be able to switch to an alternative source of water (i.
Ranney collectors) and will have I~ss flexibility in curtailing production than the Marden
WTP.
. I
- -. )
FINISHED WATER QUALITY GOALS
, j
The water quality objectives for the Columbia WTP are presented in Table 1.3 (Primary
Water Quality Goals) and Table 1.4 (Secondary Water Quality Goals).
- .J
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
c --
- 1
- ")j .~ .
771
- ,- .
- J
, j~ .~ j...
1 ,400
200
...J
c: 1 ,000
+:;
n:s
C1)
::s
a.ij 800
C1)
n:s
c:c
800
600
600
400
200
Pilot Study:
Membranes with
OAF Pretreatment
Pilot Study:
Membranes without
Pretreatment
'o:::::t .
:::J
c::(
m ,
C'.J
-:
a5-
U)
T"'-
+-'
Cl..
Marden WTP Intake
- .. -
Highway 21 Bridge (Columbia WTP Intake)
,---............. .........,--,--,--,--,--
'o:::::t
,--
'o:::::t
,--,--\ /,--,--,--,--,--,--,--,--
'o:::::t
,--
Q")
,--,--,--
Q")Q")Q")
,-- ,--,--
Date
BOISE RIVER ALGAE ENUMERATION
Figure 1.
Table 1.Primary Treated Water Quality Goals
Columbia Water Treatment Plant
United Water Idaho
Particulate/Microbial
Turbidity
Giardia
Virus
Cryptosporidium
-c::: 0.3 NTU 950/0 of time
3 log removal/inactivation
4 log removal/inactivation
3 -5 log removal/inactivation a
. ,
. I
, j
OBPs
THMs (SOS)
HAAs (SDS)
BrO3
CIO2a Depending on Risk Assessment Resultsb May result form ozonation or on-site chlorine generation
64 Jlg/L (80% of MCL LRA)
48 Jlg/L (80010 of MCL LRA)
8 Jlg/L b
8 mg/Lc
Table 1.Secondary Treated Water Quality Goals
Columbia Water Treatment Plant
United Water Idaho
Tastes & Odors
ColorFe
No objectionable odor
-c:::3ACU
-c::: 0.3 mg/L
-c::: 0.05 mg/L
. i
SUMMARY
,~ J
The results of the water quality evaluation show that the Boise River, the raw water
supply for the proposed Columbia .WTP, is a high quality source, which' allows selection
of an efficient and cost-effective treatment process. The treatment process objectives
encompass the current and anticipated federai and state water quality regulations.
- J
. - j
G:\United Water\IDAHO\6232a.OO\Correspondence\EM-OO1.doc
- ;
" J
- -.
-' jO
- I
- _. -.
TECHNICAL MEMORANDUM NO.
- - ~- !. -
EC'L 'AN'U M ' N
, ,. ,' ', ~-
((11) MONTGOMERY WATSON
carOLLO., e
UNITED WATER ID AH 0 - COLUMBIA WATER TREATMENT PLANT
To:Greg Wyatt
Dan Brown
Reference:6232A.
From:Matthew Marshall
Silas Gilbert
Date:December 21 2001
Final
-- ~
" J
Subject: Workshop No.2 - Treatment Technology Evaluation
. )
INTRODUCTION
- 1
This technical memorandum presents the results of a technology evaluation workshop
held in Botse, Idaho at the office of Carollo Engineers on February 1 , 2001. The
workshop objective was to screen a comprehensive list of available technologies to
obtain a short list of two or three candidate processes, at most, which are appropriate
for the proposed Columbia WTP. Appropriate processes are those capable of treating
the raw water quality to meet the finished water objectives defined in Workshop No.
Attendees:
,.
Bob Raczko
Dan Brown
Scott Rhead
Bill Carr
Bob Adams
Joe Jacangelo
Dan Askenaizer
Si Gilbert
Bill Lynard
Gil Crozes
Bryant Bench
Matthew Marshall
Chris Cleveland
Susumu Kawamura
Monty Marchus
United Water Resources
United Water Idaho
United Water Idaho
United Water Idaho
United Water Idaho
Montgomery Watson
Montgomery Watson
Montgomery Watson
Montgomery Watson
Carollo Engineers
Carollo Engineer~
Carollo Engjneers
Carollo Engineers
Kawamura Water Engineering
Idaho DEQ
. . j- ;
G :\U n ited Water\! DAHO\6232a. OO\Co rrespondence \EM-OO2. d oc
. !; ,
. J
. 1
, ,.~ -- ,, j
TECHNICAL EVALUATION CRITERIA
Raw water quality and finished water objectives determine process selection. The raw
water quality and finished water objectives presented in Technical Memorandum No.
include the following key parameters, which can significantly affect process selection:
Turbidity
Tastes and Odors
Algae
Total Organic Carbon (TOC)
Temperature
Microbial Removal and Disinfection
Disinfection By-Products (DBPs)
The common denominator in all municipal water supply processes is filtration. The two
filtration options evaluated herein are granular media filtration and membrane filtration.
Other processes, such as pretreatment and disinfection, are evaluated as an integrated
design in support of the filtration process.
Turbidity
Turbidity is a significant parameter in establishing the level of pretreatment required for
filtration. Each type of filter (granular media or membrane) has a limited capability to
remove solids. If raw water turbidity exceeds the solids removal capacity of the filter.
then pretreatment is required. The characteristics of granular-media and membrane
filtration related to raw water turbidity are described below.
Granular Media Filtration
The frequency distribution of Boise River turbidity with the limitations of types of granular
media filtration superimposed is presented in Figure 2.1. Comparison of the raw water
turbidity and the capabilities of different filtration processes results in the following
conclusions:
In-Line Filtration. In-line filtration means that flocculation occurs within the filter
media without previous conditioning other than chemical coagulation (flash-mix).
Based upon standard refer~nces , the maximum raw water turbidity capability of
this filtration type is 5 NTU. The Boise River exceeds this turbidity level
approximately 25 percent of the time. Therefore, in-line filtration is not considered
a reliable or feasible treatment for the Columbia wfP.
G: \U n ited Wate r\1 DAHO\62 32a. OO\Corresponde nee \EM-OO2. doe
C ~
- 1
- .~ .. !
. i.
. J
. J
::)
I-
:::-"
:c 121.0
98 6-98 9-98 12-98 3-99 6-99 9-99 12-99 3-00 6-00.
Date
BOISE RIVER TURBIDITY
Figure 1.
. ;. "
. i
, I
. )
100%
. !
". 1
Q) 60%
(.)......
::z
......(.)
40%a..
.? .. )
. I
20%
. 1
, j
c, j
. -
80%
10 12 14
Turbidity (NTU)
BOISE RIVER TURBIDITY
Figure 1.
_H -
c )
. I
1. J
Direct Filtration. Direct filtration means that flocculation follows coagulation as a
pretreatment step. This allows better conditioning of solids prior to removal within
the filter media. Based upon standard references, the maximum raw water
turbidity capability of direct filtration is 15 NTU. The Boise River raw water turbidity
is within this limit more than gg percent of the time. Therefore, direct filtration is
considered a reliable and reasonable filtration process, with respect to turbidity
removal, for the Columbia WTP.
Two Stage Filtration. Two stage filtration uses a first-stage, roughing filter for
flocculation and some solids removal prior to the second-stage, polishing filter.
Because there is. solids removal in addition to the final filtration step this type of
filtration has a maximum raw water turbidity capability of 50 NTU, based upon
standard references. This process is reliable and feasible for the Columbia WTP
with respeCt to turbidity removal.
Conventional Treatment. Conventional treatment provides flocculation and
clarification (Le., solids removal) upstream of the filter. Conventional treatment
includes such categories as lime softening, horizontal flocculation and
sedimentation, and solids contact clarification (e., the Marden wfP
Superpulsators). Based upon standard references , this process is acceptable up
to 1 000 NTU raw water turbidity and is reliable and feasible for the Columbia
wTP.
Of the three granular media filtration processes , described above , which are feasible in
terms of Boise River raw water turbidity levels, direct filtration is the least costly. It is
estimated that conventional treatment construction costs are approximately 20 percent
higher than direct filtration. Two-stage filtration construction costs fall between
conventional and direct filtration. However, two stage filtration has several inherent
disadvantages including:
Significantly higher volume waste stream for recycling with lessconcen~rated-
solids than conventional processes.
Less robust and flexible in dealing with some of the parameters of concern (e.
tastes and odors, and algae) than conventional processes.
Less flexible than conventional treatment in dealing with future regulations andunknown contaminants.
In summary, both direct filtration and conventional treatment are considered feasible
and reasonable for removal of turbidity from the Boise River. Although two stage
filtration is feasible for treating the Boise River turbidity it is not considered reasonable
and -in-line filtration is not considered feasible.
G: \U n ited Water\IDAHO\6232a. OO\Correspondence \EM-OO2. doc
. 0 ,
, '
0 i
0 7
. i
. .
Membrane Filtration
Low-pressure membrane filtration processes , including both microfiltration (MF) and
ultrafiltration (UF), have different capabilities for removal of turbidity than granular
media. The two primary methods of operation for low-pressure membranes under
consideration for the Columbia WTP are recirculation (cross-flow) operation and dead-
end operation. The Boise River raw water turbidity frequency distribution with membrane
turbidity capabilities superimposed is presented in Figure 2.2. Observations include the
following:
The raw water turbidity in the Boise River allows operation of a membrane in
dead-end mode (at the appropriate flux) at least 99 percent of the time.
Membrane operation in recirculation mode is always feasible with the Boise River
turbidity. Generally, 50 NTU .is the upper limit for turbidity without significant
impact on membrane efficiency.
In summary, low-pressure membrane technologies are feasible for treating the Boise
River raw water turbidity.
Tastes and Odors
A summary of water quality information regarding the Boise River raw water tastes and
odors presented in Workshop No.1 includes the following:
Marden WTP operational staff report instances of tastes and odors occurring in
the spring or fall, resulting in some consumer complaints.
The taste and odor events are of short dur~tion, typically two or three weeks at the
longest.
The presumptive taste and odor compounds are geosmin and Mia.
Powered activated carbon (PAC) is used for taste and odor treatment at the
. Marden WTP.
The maximum dose of PAC fed at the Marden wfP is limited by subsequent
removal of PAC on the filter in the direct filtration mode. Removal of PAC is the
limiting factor in taste and odor control in this case.
The following technologies were considered for treatment of tastes and odors at the
roposed Columbia wfP:
Potassium permanganate is effective in treating some tastes and odors;however
it generally ineffe~tive in dealing with geosmin and MIB. In some cases, it
increases tastes and odors resulting from algae because of the effect of this
strong oxidant in rupturing the cellular membrane.
Ozone is typically effective in removing geosmin and MIB; however, the capital
cost for installation of ozone is higher than other available technologies. Typically,
it is not cost-effective for removal of tastes and odors that only occur seasonallyfor short durations:
G :\U n ited Water\1 DAHO\6232a. OO\Correspondence \EM-OO2. doc
" -, ... ~- ,. ~. j
. i
. J
. ;.,..
- J
PAC is generally effective for removal of geosmin and MIB; however, PAC must
be subsequently removed by the treatment process (e., through clarification or
filtration). PAC has a low capital cost compared to ozone but a relatively high
operating cost. It can be cost-effective for seasonal taste and odor problems ofshort duration.
Granular activated carbon (GAC) is effective in removing geosmin and MIB.
Typically, GAC is installed as a granular filter media; it is more costly when
installed as a stand alone GAC contactor. Typically, GAC is cost-effective in
treating long term taste and odor problems.
In summary, potassium permanganate is eliminated from further consideration because
it is not effective for the taste and odor causing compounds in the Boise River water
supply. Ozone is eliminated from furth~r consideration because its high capital costs
make it unsuitable for seasonal , short-term taste and odor events. GAC is considered
economical and feasible; however, only as a filter media and not as a separate post-
filter adsorber. PAC is considered the most appropriate taste and odor control process
for seasonal short duration events. However, conventional treatment is required to
. provide removal of PAC (at sufficiently high doses) to ensure adequate taste and odor
control capability.
. Algae
Limited quantitative algae data are available for the Boise River. One early sample taken
April 12 , 1990, shows algae levels (2 900 organisms per ml) are approaching the limits
of direct filtration (3 000 organisms per ml).
Weekly algae samples were collected from July 24 through October 16, 2001 at both the
Marden WTP intake and the proposed Columbia WTP intake (Highway 21 bridge
location). Algae levels ranged from 139 to 1 310 organisms per ml at t~e Marden WTP
intake and 141 to 1 570- organisms per ml at the proposed Columbia VVTP intake.
Marden WTP staff reported that algae production peaked by the end of June in 2001
prior to sampling.
The Marden WTP operations staff report that algae can be a moderate to serious
problem on occasion. Plant capacity in a direct filtration mode has been constrained by
. filter clogging algae. The conventional treatment process at the Marden WTP allows for
efficient removal of algae when required. It is recommended that an algae removal
pretreatment process is also included at the Columbia WTP.
Several conventional pretreatment technologies were evaluated for algae removal.
These processes include the following:
Dissolved air flotation is one of the best algae removal technologies because it
takes advantage of the lightweight and buoyant characteristics of most algae.
Algae are more readily removed by flotation than by gravity settling.
Upflow, solids contact clarification (e:g., Superpulsator) effectively removes algae
G: \U n ited Water\1 DAH 0\6232a. OO\Correspondence \EM-OO2. d oc
. ,
by entrapment in the sludge blanket. This is the existing pretreatment process at
the Marden WTP.
High rate sand-ballasted sedimentation (e., Actiflow) can remove algae but not
as effectively as dissolved air flotation.
Rectangular, horizontal flocculation and sedimentation basins are not as effective
for algae removal and uncovered basins can contribute to the problem by
encouraging growth of algae.
In summary, a pretreatment process is recommended for removal of algae before
filtration. The best technologies are dissolved air flotation or solids contact clarification
based on effectiveness and efficiency.
TOC Removal
The Soise River is a high quality source with respect to total organic carbon (TOC).
Episodes of elevated TOC appear to be associated with the flushing of the river .by high
flows at the beginning of the irrigation season. Typically, high TOC events are less than
a week in duration. The 50th percentile TOC is 1.7 mg/L.; the 95th percentile is less than
5 mg/L.
. j
Current regulations do not require a direct filtration process to remove TOC as an
enhanced coagulation requirement. In addition, based upon the review of THM
information potential presented in Technical Memorandum No., it does not appear that
enhanced coagulation for removal of disinfection by-product precursors will be required.
It appears from available data that the proposed Columbia wfP will meet DBP
requirements without enhanced coagulation while using free chlorine as a distribution
system disinfectant.
. J Disinfection of Microbial Contaminants
. 1
; J
The following finished water quality objectives were recommended in Technical
Memorandum NO.1 for removal or inactivation of microbial contaminants:
,- i
Virus
The virus removal/inactivation objective is 4-logs:
Direct filtration. virus removal credit is 1-log; therefore, an additional 3-log virus
inactivation through disinfection is required. Worst case CT for 3-log inactivation
with free chlorine is 9 mg/L8minutes at C.
Conventional treatment virus removal credit is 2-log; therefore, an additional 2-
log virus inactivation through disinfection is required. Worst case CT for 2-log
inactivation with free chlorine is 6 mg/Leminutes.at C.
Membrane filtration virus removal credit may range from no credit to
G :\U n ited Water\1 DAHO\6232a. OO\Corresponde nee \EM-OO2. doe
. i
. ..
log removal credit; therefore , the virus inactivation requirement may range
from 1-109 to 4-log. To achieve this, CT with free chlorine may range from 3 to 12
mg/L-minutes at 0.
Giardia
Giardia cyst removal/inactivation objective is 3-logs:
Direct filtration Giardia cyst removal credit is 2-log; therefore, an additional1-log
Giardia inactivation through disinfection is required. Worst case CT for 1-log
inactivation with free chlorine is 81 mg/L-minutes at 0.C and pH 7.4 (90th
percentile pH).
Conventional treatment Giardia cyst removal credit is 2.log; therefore, an
additional 0.log Giardia inactivation through disinfection is required. Worst case
CT for 0.5 log inactivation with free chlorine is 41 mg/Leminutes at 0.C and pH4 (90th percentile).
Membrane filtration Giardia removal credit may be as high as 3-log; therefore
additional Giardia inactivation credit by disinfection may not be required. In this
case, virus CT governs for free chlorine disinfection following membrane
filtration.
Cryptosporidi urn
Cryptosporidium oocyst removal requirement will be 2-logs under the Interim Enhanced
Surface Water Treatment Rule (IESwrR). This credit is available by meeting the
turbidity requirement of the IESwrR (Le., less than 0.3 NTU filter effluent 95 percent of
time; less than 1 NTU combined filter effluent 100 percent of the time).
. The Long Term (2) ESwrR will require 24 months of raw water quality monitoring to
. classify the C1jtptosporidium source water concentration range in "action bins
Additional removal or inactivation requirements (above the 2-logs removal required
under the IESwrR) are dependent on water quality characterization , ranging up to an
additional 2.5:-log requirement. Therefore, the total removal/inactivation requirement for
Cryptosporidium will range between 2 and 4.logs.The foliowing issues are pertinent in
evaluating technology for Cryptosporidium inactivation at the Columbia wrp:
- ). )
- 1;
Chlorine is ineffective for illdctivation of Cryptosporidium.
The Cryptosporidium inactivation CT requirement for ozone and chlorine dioxide
is unreasonably high at low temperatures. For example, the worst case
Cryptosporidium CT with ozone is 45 mg/L-minutes for 2.log inactivation at
The lowest cost technology for inactivation of Cryptosporidium is ultraviolet (UV)
disinfection.
In summary, free chlorine is an effective and economical disinfectant for virus. The
contact time for virus inactivation with free chlorine is low; typically, inactivation can
occur within a channel or pipeline without construction of a separate CT basin. In
addition, free chlorine will be required as the secondary disinfectant for the Columbia
WTP; that is , to maintain a distribution system disinfectant residual.
G: \U n ited Water\IDAH O\6232a. OO\Co rrespondence \EM-OO2. doc
UV disinfection is recommended as the most cost-effective method of providing the
required inactivation of Cryptosporidium. Cryptosporidium oocysts are resistant to
chlorine and require excessive contact time and dose for disinfection by ozone and
chlorine dioxide under worst case (low temperature) conditions. UV disinfection will also
provide more than the required inactivation of Giardia cysts at the same operating
dosage to provide required disinfection of Cryptosporidium. UV may not be cost-
effective for inactivation of some virus.
- -
- I
CONCLUSIONS OF TREATMENT TECHNOLOGY EVALUATION
Table 2.1 presents a summary of treatment technologies evaluated in Technical
Memorandum No.
Ois~olved air flotation (OAF) pretreatment is recommended for removal of algae
PAC (when used for taste and odor control), and reducing solids loading to the
filtration process to increase the filtration rate, or membrane flux.
UV disinfection is recommended for inactivation of Cryptosporidium oocysts and
Giardia cysts. Free chlorine is recommended for inactivation of virus and as a
distribution system disinfectant.
. j- ,
- 1 The recommended filtration process is either granular media filtration or
membrane filtration.
These alternatives form the basis for two separate treatment options:
1. Pretreatment (OAF) followed by granular media filtration followed by UV disinfection.
. 2. Pretreatment (OAF) followed by me~brane filtration followed by UV disinfection.
The selection of the final treatment process between the granular media and membrane
filtration options is based upon refined design criteria and cost estimates as presented in
Workshop No.3 and documented in Technical Memorandum'No. 3.
G: \U n ited Wate r\1 DAHO\6232a. OO\Co rrespo nde n ce \EM-OO2 .doc
O_
-
,
~
"
.
_
"
O
l.
.
--
-
--
.
.
""
.
.
~
~
-
'
-'
-
"
"
'-
-
-
-
-
-
""
,
~
-
,
,
"
,
-
_
o
'..
~.
_
,
,
-
,.
.
-
-
-
-
-
.
...
.
-
-
,
.
.-
.
.
,-
-
,
.
-
~
TA
B
L
E

2
.
SU
M
M
A
R
Y

O
F

T
R
E
A
T
M
E
N
T

T
E
C
H
N
O
L
O
G
Y

E
v
A
L
U
A
T
I
O
N
Ra
w

Wa
t
e
r

Pa
r
a
m
e
t

r
s
1\
,
0
.J
.
~
"'
~
~O
)

'"

~

(j
\

~
~
.
...
0~

~~
v
Re
m
a
r
k
s
Pr
e
t
r
e
a
t
m
e
n
t
Di
s
s
o
l
v
e
d

A
i
r

F
l
o
t
a
t
i
o
n
So
l
i
d
s

C
o
n
t
a
c
t

C
l
a
r
i
f
i
e
r
Sa
n
d
-
ba
l
l
a
s
t
e
d

C
l
a
r
i
f
i
e
r
Ho
r
i
z
o
n
t
a
l

F
l
o
c
/
S
e
d
Gr
a
n
u
l
a
r

M
e
d
i
a

F
i
l
t
r
a
t
i
o
n
Co
n
v
e
n
t
i
o
n
a
l
Tw
o

S
t
a
g
e
Di
r
e
c
t
In
-
li
n
e
Me
m
b
r
a
n
e

F
i
l
t
r
a
t
i
o
n
Mi
c
r
o
f
i
l
t
r
a
t
i
o
n
Ul
t
r
a
f
i
l
t
r
a
t
i
o
n
Ta
s
t
e

a
n
d

O
d
o
r

C
o
n
t
r
o
l
GA
C
PA
C
Oz
o
n
e
Po
t
a
s
s
i
u
m

P
e
r
m
a
n
g
a
n
a
t
e
Di
s
i
n
f
e
c
t
a
n
t
s
.
U
l
t
r
a
v
i
o
l
e
t
Oz
o
n
e
Ch
l
o
r
i
n
e

D
i
o
x
i
d
e
Ch
l
o
r
i
n
e
Le
g
e
n
d
Sh
a
d
i
n
g

D
e
n
o
t
e
s

N
o
t
Ap
p
l
i
c
a
b
l
e

t
o

t
h
e
Pa
r
a
m
e
t
e
r
Mo
s
t

e
f
f
e
c
t
i
v
e

p
r
o
c
e
s
s

f
o
r

a
l
g
a
e

r
e
m
o
v
a
l
Go
o
d

p
r
o
c
e
s
s

f
o
r

a
l
g
a
e

r
e
m
o
v
a
l
Ac
c
e
p
t
a
b
l
e

"
p
r
o
c
e
s
s

f
o
r

a
l
g
a
e

r
e
m
o
v
a
l
Ac
c
e
p
t
a
b
l
e

p
r
o
c
e
s
s

f
o
r

a
l
g
a
e

r
e
m
o
v
a
l
Mo
s
t

"
ef
f
e
c
t
i
v
e

p
r
o
c
e
~
s

f
o
r

a
l
g
a
e

a
n
d

T
&
O

(
P
A
C
)
Ma
r
g
i
n
a
l

p
r
o
c
e
s
s

f
o
r

a
l
g
a
e

a
n
d

T
&
O

(
P
A
C
)
Re
q
u
i
r
e
s

p
r
e
t
r
e
a
t
m
e
n
t

f
o
r

a
l
g
a
e

a
n
d

T
&
O

(
P
A
C
)
No
t

a
c
c
e
p
t
a
b
l
e

f
o
r

t
u
r
b
i
d
i
t
y

r
e
m
o
v
a
l
Re
q
u
i
r
e
s

p
r
e
t
r
e
a
t
m
e
n
t

f
o
r

a
l
g
a
e

a
n
d

T
&
O

(
P
A
C
)
Re
q
u
i
r
e
s

p
r
e
t
r
e
a
t
m
e
n
t

f
o
r
a
l
g
a
e

an
d

T
&
O

(
P
A
C
)
Lo
w

c
a
p
i
t
a
l

c
o
s
t
Lo
w
e
s
t

c
a
p
i
t
a
l

c
o
s
t
,
H
i
g
h
e
s
t

ca
p
i
t
a
l

c
o
s
t
No
t

e
f
f
e
c
t
i
v
e

f
o
r

a
l
g
a
e
-
r
e
l
a
t
e
d

T
&
O
Lo
w
e
s
t

o
v
e
r
a
l
l

c
o
s
t

f
o
r

Cr
y
p
t
o

an
d

Gi
a
r
d
i
a
Hi
g
h

o
v
e
r
a
l
l

c
o
s
t

f
o
r

Cr
y
p
t
o

an
d

Gi
a
r
d
i
a
Hi
g
h

o
v
e
r
a
l
l

c
o
s
t

f
o
r

Cr
y
p
t
o

an
d

Gi
a
r
d
i
a
Lo
w
e
s
t

o
v
e
r
a
l
l

c
o
s
t

f
o
r

v
i
r
u
s
Hi
g
h
e
s
t

r
a
n
k
i
n
g
Be
t
t
e
r

t
h
a
n

a
v
e
r
a
g
e

r
a
n
k
i
n
g
Ne
u
t
r
a
l
/
a
v
e
r
a
g
e

r
a
n
k
i
n
g
No
t

f
e
a
s
i
b
l
e
/
l
o
w
e
s
t

r
a
n
k
i
n
g
""
"
~
,
~
.
.
~
,
~
,
,
l_
.
.
_
.
.
.
-
"~
.
~
"
.,
,
.
t!
j
'"
.
~,

.
.
_
"
"
,
,
""
"
~
.
.
"
.
.,.
,
;
w-
.
_
~
'
.
"
'
-
'
--
-
-
.
.
"
.
"
"
.
-
.
.
,.
.
"
"
"
-
~
,
,
...
-
e H N. ~ CA . M' E, M 0 R N D N O.
MONTGOMERY WATSON ~.H '79 .!-.!-..
~ ~
UNITED WATER IDAHO - COLUMBIA WATER TREATMENT PLANT
To:Greg Wyatt
Dan Brown
Reference:6232A.
From:Matthew Marshall
Silas Gilbert
Date:December 28, 2001
Final
, J
Subject: Workshop No.3 - Process Selection
, !
INTRODUCTION
This technical memorandum presents the results of a process selection workshop held
in Boise, Idaho at the office of Carollo Engineers on February 23, 2001. The workshop
objective was to select the treatment process for the proposed Columbia Water
Treatment Plant (WTP). The two short-listed process alternatives from Workshop No.
were as follows:
Pretreatment followed by granular media filtration and UV disinfection.
Pretreatment followed by membrane filtration and UV disinfection.
. .
Planning level design criteria and 'cost estimates for the two short listed alternatives
were presented and reviewed during the workshop.
. J
Attendees:
Greg Wyatt
Dan Brown
Scott Rhead
Bill Carr
Bob Adams
John Dyksen
Bob Raczka
Si Gilbert
Garry Wohlgemuth
Gil Crozes
Bryant Bench
Matthew Marshall
Chris Cleveland
United Water Idaho
United Water Idaho
United Water Idaho
United Water Idaho
United Water Idaho
. United Water Resources
United Water Resources
Montgomery Watson
Montgomery Watson
Carollo Engineers
Carollo Engineers
Carollo Engineers
Carollo Engineers
D:\Si's 0 Drive\EM-OO3.doc
REVIEW OF WATER QUALITY AND PROCESS SCREENING
Water quality issues and the preliminary screening of treatment process alternatives
during Workshops No.1 and 2 were reviewed briefly. Additional comments and
clarification from the Marden WTP operational staff were provided.
. ~
Turbidity
Plant staff confirmed that the highest Boise River raw water turbidity
observed at the Marden WTP is 17 NTU.
. - -- !
. i
Turbidity spikes at the new raw water supply intake structure for the
Columbia WTP are attributed to recirculation flow from the intake pumps
(opera~ed by Micron Technology) stirring up sediment. This is a short-term operational issue which will not occur when the Columbia WTP is
constructed and operating.
Raw water turbidity analysis presented in Workshop No.1 represents theworst-case condition because:
recirculation of water and sediment will not occur when pumping
to the water treatment plant, and
turbidity at the new intake (closer to Lucky Peak Reservoir) is
generally lower than turbidity at the Marden intake (six miles
downstream).
- 1
Total Organic Carbon (TOG)
The higher values of TOC (greater than 2.5 mg/L) are normally due to thefirst flush" at the beginning of the irrigation system.
High TOC events are limited to several weeks duration.
Taste and Odor
: J
Taste and odor events are seasonal and of short duratron (several weeksin length).
Taste and odor is treated at the Marden WTP using powdered activatedcarbon (PAC).
. j
Use of PAC at the Marden WTP is limited during direct filtr~tion
operation; however, the Marden WTP has the option of using solids
contact clarifiers to accommodate high doses of PAC during taste andodor even~.
- ,
Algae
There have been occurrences of algae in the Boise River, which havelimited the capacity of the Marden WTP when operated in a direct
filtration mode.
Algae enumeration data collected in the Summer and Fall of 2001 showhigher algae levels at the Columbia WTP intake (Highway 21 Bridge)than at the Marden WTP.
D:\Si's 0 Drive\EM-OO3.doc
Previously, the potential use of the Columbia WTP as a source of water for aquifer
storage and recovery (ASR) projects has been considered. In some circumstances
production for ASR affects process selection. For example, the specific flux of a
membrane decreases with temperature while granular media filtration capacity is
undiminished. After discussion , the following are decisions and discussions during
Workshop No.
. ,. ,
, 1
. ). .
. J
Aquifer Storage and Recovery
The initial maximum production for ASR required from the Columbia WTP will be
to 2 mgd.
Excess capacity of 1 to 2 mgd for ASR is available using either a granular media
filtration or membrane filtration option.
ASR capacity of 2 mgd will be provided at the Columbia WTP.
The Micron ASR concept is based upon elimination of bacteria without the use of
free chlorine. UV disinfection provides the same benefit for the Columbia WTP.
A potential use of the Columbia WTP is for treatment of groundwater for
reinjection under ASR, with the following benefits:
Arsenic can be removed through coagulation and stored for summer
reuse.
Treatment of groundwater in the winter will raise the temperature of the
water supply and increase membrane flux.
- .
Based upon the capacity requirements selected, ASR production is not a criteria for
selection of the treatment process options at the Columbia WTP.
Process Alternatives
The two process alternatives selected in Workshop No.2 for further consideration were:
Dissolved air flotation pretreatment + granular media filtration + UV disinfection
Dissolved air flotation pretreatment + membrane filtration + UV disinfection
Dissolved air flotation (OAF) pretreatment and UV disinfection are common to both
process alternatives.
OAF Pretreatment
The existing Marden WTP process includes conventional pretreatment in the form of
upflow clarifiers (Superpulsators). Historically, the Marden WTP clarifiers have been
used to treat short-term events of elevated turbidity and TOC (color).
The pretreatment selected for the Columbia WPT should fill the same role as the
existing pretreatment at the Marden WTP. The recommended pretreatment process for
the Columbia WTP is OAF. Advantages of OAF include the following:
D:\Si's D Drive\EM-OO3.doc
. ," .. j, j.
. J
,. i
u J
, ".~. ,' "
" 1
, J
. 1
;, ,~' ,
. J
", j
The lightweight nature of the constituents removed (e.algae, PAC, andcoagulant) indicate that a flotation process is more effective than a gravity settling
process.
OAF operates at higher rates than most comparable alternative clarification
processes resulting in a smaller footprint requirement:
Conventional rectangular sedimentation basins: 0.75 gpm/sq. ft.
Marden WTP Superpulsators: 2.37 gpm/sq. ft.
Proposed Columbia WTP OAF: 6 gpm/sq. ft or greater (subject to pilot
plant test verification).
OAF requires a lower coagulanJ dose than other high rate clarification processes.
OAF is not upset by the water temperature variation that occurs in the Boise
River.
Quick startup time: 45 minutes for OAF, compared to other clarificationprocesses which can take 8 to 24 hours for stable operation.
In summary, dissolved air flotation is the best pretreatment for Boise River raw water
quality: algae, low turbidity and absence of heavy solids. Most settleable solids (clays
and silts) are removed naturally in the three upstream impoun~ments (Lucky Peak
Reservoir, Arrowrock Reservoir, and Anderson Ranch Reservoir). It is proposed that
OAF pretreatment will increase the capacity of the downstream filtration process. The
following granular media maximum filtration rates are proposed:
5 gpm/sq. ft. without pretreatment
0 gpm/sq. ft. with pretreatment
The difference in filtration rates means that a plant with a 10 mgd nominal capacity
without pretreatment can operate at 12.3 mgd with pretreatment.
Granular Media Filtration Option
The proposed design criteria for the granular media filtration option are shown in Table
, assuming a 6 mgd plant capacity. The design criteria also include pretreatment
(OAF) and disinfection (UV) which are common between both granular media and
membrane filtration options.
The process flow diagram for the granular filtration process is shown in Figure 3.1. Theproposed process flow diagram for the Columbia WTP is very similar to the existing
Marden WTP (Phase 1) for treatment of the Boise River. The proposed Columbia WTP
granular media filtration option is best reviewed in comparison to the existing MardenWTP.
D:\Si's 0 Drive\EM-OO3.doc
River Intake and Raw Water Pump Station
These facilities are not included in the process criteria or alternatives evaluation. The
same river intake facilities are used for either the granular media or membrane filtration
alternatives.
,.-. ~
. J
,:d
- J Flash Mix
- -
. I
The pump diffusion flash mix recommended for the Columbia WTP is similar to the
installation at the existing Marden WTP.
Flocculation
, 1
The criteria used for flocculation are similar to the criteria for the direct filtration process
train at the Marden WTP (Phase 2). Two flocculation basins are proposed to allow
- maintenance. Criteria include two-stage flocculation with 20 minutes total detention time.
The flocculation basins will be used at all times either in a direct filtration mode or in a
dissolved air flotation mode.
DAF Basins
. J
Two flotation basins are proposed to allow maintenance. The assumed surface loading
rate is 4 gpm/sq. ft. Pilot testing will establish the final design criteria for the surfaceloading rate.
, !! -
Filters
Five filters are proposed with a filtration rate of less than 8 gpm/sq. ft. when one filter is
out of service for backwash and less than 6.5 gpm/sq. ft. when all filters are in service.
The proposed filter media design is more robust (i.more depth and solids storage
capacity) than the Marden WTP design to allow more efficient operati~n in a direct
filtration mode. The proposed media includes 24 inches of GAG, 30 inches of anthracite
coal, and 10 inches of silica sand for a total media depth of 64 inches. The Ud ratio of
the proposed design is 1600, compared to the Marden WTP Ud ratio of 1185..
. _
Filter Backwash
. 1 The proposed filter backwash facilities are similar to those at the Marden WTP and, include:
. .
Surface Wash
Air Scour
Pumped Backwash
A filter waste washwater flow equalization basin is proposed; however, this facility may
not be required if the treatment plant site al!ows construction of the solids handling
lagoons at a elevation which can receive the filter waste washwater flow by gravity.
D:\Si's 0 Drive\EM-OO3.doc
; 1
. 1
, ".. j- j" .. ." j
. i
. .
Disinfection
UV disinfection will provide a multiple barrier for inactivation of Giardia and
Cryptosporidium. Free chlorine disinfection will provide a multiplE? barrier for inactivation
of virus.
Solids Handling
Lagoons are the recommended facilities for solids handling. The proposed lagoon
criteria are similar to those used at the Marden WTP; however, the size of the lagoons
are slightly smaller. Two lagoons are proposed with annual solids loading rate of
approximately 4.5 Ibs/sq. ft. per year (dry solids). Lagoons will provide the following
functions at the Columbia WTP:
Filter waste washwater equalization (depending upon site constraints).
Treatment (clarification) of filter waste washwater prior to recycling.
Storage of solids removed from the treatment process.
Solids thickening.
Solids dewatering.
Chemical Feed
Facilities proposed for the Columbia WTP include the following:
Ferric chloride as the primary coagulant
Cationic polymer (PEC) as a coagulant aid
Anionic/nonionic polymer (PEA) as a flocculant aid and filter aid
Sodium hypochlorite as a disinfectant
Caustic soda for pH adjustment and corrosion control
Corrosion inhibitor
Membrane Filtration Option
The proposed design criteria for the membrane filtration options are shown in Table 3.
The design criteria also include criteria for OAF pretreatment and UV disinfection , which
are common between both granular media and membrane filtration options.
The process flow diagram for the membrane filtration option is shown in Figure 3.
Most process elements of the membrane filtration option are the sam!3 as those fa! thegranular media filtration option.
River Intake and Raw Water Pump Station
These facilities are not included in the process criteria or alternatives evaluation. The
river intake facilities used for either the granular media or lTIembrane filtration
alternatives are essentially the same; however, if the OAF pretreatment is deleted or
deferred , the raw water pump station can also serve as the membrane feed pump
station.
D:\Si's 0 Drive\EM-OO3.doc
. ..,
Flash Mix
- 1 The same pumped diffusion flash mix is recommended for both the granular media and
membrane filtration options.
. ~
Flocculation
, -
The same flocculation criteria are recommended for both the granular media and
membrane filtration options.
DAF Basins
The same DAF basin criteria are recommended for both the granular media and
membrane filtration options.
Membrane Filters
: ,
Membrane capacity is temperature dependent because increasing viscosity decreases
specific flux. The proposed membrane criteria is for a capacity of 6 mgd at a water
temperature of 10oC (500 F). Data presented in Technical Memorandum No.1 indicate
the temperature is less than 500F approximately 70 percent of the time.
- 1
- -
A conservative rule of thumb is that membranes lose 2.5 percent specific flux capacity
with every 1OC decrease in water temperature. Using this assumption, the proposed
membrane filtration facility has a minimum (wintertime) capacity of 4.5 mgd. In practice
however, the 6 mgd capacity can be maintained year round through use of DAF
pretreatment, as necessary to reduce fouling potential, or the use of increased chemical
cleaning frequency.
. J
Membrane flux is not shown in the design criteria because this parameter is specific to
individual membranes.
, j
Filter Backwash
. .
The proposed membrane backwash facilities include pumped backwash and a flow
equalization basin for membrane waste washwater. Depending upon the site
topography, it may be possible to use the solids handling lagoons for equalization and
eliminate a separate basin.
. i
:.J Disinfection
- J
The same criteria for UV disinfection (inactivation of Giardia and Cryptosporidium) andfree chlorine disinfection (inactivation of virus) are used for either the granular media or
- membrane filtration alternatives.
Solids Handling
The solids produced by either the granular media filtration option or membrane filtration
option are equivalent. Therefore, the same lagoon criteria .are used for both options.
However, the amount of waste washwater produced by the membrane filtration option is
D:\Si's D Drive\EM-OO3.doc
: J
. 1
: .: !' .
. I
" ,, .
. I
~ .
greater than the washwater production of the granular media filtration option. This is
because the membrane option results in a waste stream that is 5 to 10 percent of plant
. production , whereas, the waste stream for the granular media option is 2 to 4 percent of
production.
Several potential cost-saving alternatives for disposal of membrane waste washwater
are not available for the granular media filtration option. The Idaho Department of
Environmental Quality has indicated a willingness to consider approving direct discharge
of membrane waste washwater to the Boise River during those times that chemical
coagulation is not used as a pretreatment or, the waste washwater could be used for
irrigation or industrial purposes. The costs and benefits of these options should be
considered further in predesign.
Optiqnsfor handling membrane waste washwater include the following:
Direct discharge to the lagoons. Coagulation and flocculation of membrane waste
washwater is required if coagulation is not used as a pretreatment process.
Recycling of equalized membrane waste washwater to the OAF pretreatment
Direct discharge to the Boise River. Dechlorination may be required if chlorinatedwater is used for backwash.
Direct discharge to an irrigation or industrial use in the vicinity of the plant. This
requires consideration of non-irrigation (winter) uses or winter storage.
Chemical Feed
Chemical facilities proposed for the membrane filtration option include the following:
Ferric chloride as the primary coagulant
Sodium hypochlorite as a disinfectant
Caustic soda for pH adjustment and qorrosion control
PAC
Other chemicals as required for membrane cleaning.
COST ANALYSIS
Table 3.3 contains estimated capital costs for a 6 mgd capacity Columbia WTP. Costs
are presented for both the granular media filtration option and the membrane filtration
option. The costs presented are planning level of accuracy (:t 300/0) and serve as a
comparison between the capital costs of the granular media and membrane filtration
options. The estimated costs were prepared using the proposed design criteria and unit
cost assumptions fqr process elements based on similar projects. No detai~ed planning
or quantity takeoffs are included.
D:\Si's 0 Drive\EM-OO3.doc
, J Table 3.Estimated Capital Costs
Columbia Water Treatment Plant (6 mgd)
United Water Idaho
'... '
Item
General Requirements
Civil/Site C
Inlet Structure
OAF (including flocculation)
Granular Media Filters e
Membrane Filters f
Membrane Building
Membrane Feed Pump Station.
UV Disinfection
CT/Clearwell i
Pump Station j
Chemical Feed Facility
Washwater Equalization Basins
Solids Separation Basins
Operations Building
HV AC
Instrumentation n
Electrical 0
Construction Subtotal
$/gallon/day
Eng./Contingency (300/0)
Total Capital Costs
ENR Cost Index = 4800
General requirements = 70/0 of con$truction subtotal and includes mobiliza~ion , demobilization , bondsand other Division 0 and Division 1 requirements.
Based on Marden WTP costs updated to current ENR Cost Index.
OAF costs include equipment, installation basins , and building.
Granular media filter cost = $2 500/sq. ft., including building.
Membrane purchase price = $0.33/gpd, including CIP but excluding backwash pumps and feed
pumps; installation -= $0.07/gpd, or 20% of equipment cost.
Membrane building unit cost = $11 O/sq. ft. for 5,000 sq. ft.
UV disinfection unit cost = $0.11 /gpd.
Basin unit cost = $0.60/gallon.
Pump station unit cost = $1 ,900/hp (including building); includes utility pumps C).nd backwash pumps.
Includes chemical systems and building.
Basin unit cost = $12/sq. ft.
Operations building unit cost = $150/ sq. ft. for 2 000 sq. ft.
Instrumentatipn = 6% for granular filtration; 4% for membranes.
0 Electrical = 14% for granular filtration; 10% for membranes. Costs do not include standby power.
Granular Filtration ($)
720 000
800 000
200 000
1,400 000
800 000
Membrane Option ($)
760 000
800 000
200 000
1 ,400 000
'd 1
" 1
, !
700 000
770 000
450 000
220 000
300 000
500 000
300 000
150 000
520 000
200 000
$11 030 000
$1.
310 000
340 000
2,400 000
550 000
680 000
700 000
700 000
330 000
100 000
200 000
500 000
300 000
150 000
380 000
960 000
$12 , 11 0 000
$2.
630 000
740 000
. I
- I
~ 1
, i
. ,! -
, J
D:\Si's 0 Drive\EM-OO3.doc
- i
- -
The estimated construction cost of the membrane filtration option ($12 110 000) is
approximately 10 percent higher than the estimated construction cost of the granular
media filtration option ($11 030 000). At a planning level-of cost estimating, the
difference is within the tolerance of estimating accuracy (300/0).
- 1 Table 3.4 presents estimated operation and maintenance (O&M) costs for the Columbia
WTP operating at an average day capacity of 6 mgd. The estimated unit costs of
production for granular media filtration and membrane filtration are equal.
, j
Table 3.Estimated O&M Cost Comparison ~t 6 mgd Production
Columbia Water Treatment Plant
United Water Idaho
MembraneItemGranular Filtration ($)Option ($)
GAC Replacement 000
Membrane Replacement 108 000
labor 125 000 500
Power
(g)
$0.07/kW-322 000 370 000
Chemicals 100 000 500
Maintenance 50,000 000
Annual O&M Cost $666 000 $663 000
Cost per thousand gallons $0.$0.
- J
. '
. J
, j- ;
ALTERNATIVES EVALUATION
Criteria were selected for the evaluation of the two proposed process options for the
Columbia WTP: membrane filtration or granular media filtration. Initially, 24 separate
decision criteria were identified. The six criteria with the greatest weight in process
selection are preserrted herein.
, J Reliability
. i
, I
Reliability has the highest weight in determining process selection. The issue of
reliability is broken down into three subcategories, as' follows:
Finished Water Quality
The membrane filtration option has a higher reliaqility in producing finished water
meeting the water quality objectives for turbidity and removal of microbial contaminants.
This is because membrane filtration relies only on physical removal whereas granular
media filtration water quality is dependent upon adequate chemical pretreatment and
optimization of the filtration process. Although maintaining membrane filtration capacity
D:\Si's 0 Drive\EM-OO3.doc
. j. j
may require optimization of pretreatment, the finished water quality is independent of
process optimization or raw water quality.
".._
Regulatory Performance
Membrane filtration has a higher reliability in meeting future water quality regulations
than granular media filtration. The filtered water turbidity standard has been adjusted
three times in the last decade , at each step it becomes more difficult for granular media
filters to meet the new standard. Membrane 'filtration is a more reliable process for
. meeting future water quality regulations governing turbidity and particulate removal
because it acts a physical barrier.
. '-, j
, 1
Reliability During Unmanned Operation
One of the objectives for the Columbia WTP is unmanned operation with remote
surveillance. Membrane filtration is more reliable, in terms of water quality, for remote or
unmanned operation because finished water quality is independent of pretreatment or
process optimization. Granular media filtration quality is dependent upon factors such
proper operation of chemical feed systems, flash mix, and flocculation, all optimized for
specific water quality conditions. If water quality changes or chemical feed is interrupted
the granular media finished water quality can be adversely affected.
.. I
. j. )
. i Capital Costs
As presented in Table 3., the estimated capital cost of membrane filtration is slightly
higher (approximately 10 percent) than the estimated capital cost of granular media
filtration; however, within a planning level of accuracy they are equivalent. The cost
estimates presented herein do not contain sufficient detail to evaluate all differences
between the two process options. For example, the membranes can be constructed as
an at-grade facility without the excavation required for construction of granular media
filters. Depending upon the site selected and the depth to basalt rock
, -
this could
represent a significant cost advantage to membrane filtration.
. I
~ j
L.1
Operation and Maintenance Cost
As shown in Table 3.4, the O&M costs for granular media filtration and membrane
filtration options are equivalent. However, there is a difference between the two options
in that granular media filtration has a h~gher labor cost component. Membrane filtration
allows a reduction in labor with reliable unattended operation. The membrane O&M
labor cost savings are offset by the requirement to replace the membrane filters over
ti~e. If the membrane replacement cost ~an be treated as a depreciated cost capital
this may provide some advantage for the membrane filtration option.
. .
Modularity for Expansion
- )
The membrane filtration option allows expansion in smaller increments than the granular
media option. As an example, the Marden WTP was originally constructed with four
D:\Si's 0 Drive\EM-OO3.doc
" i
, U"
- i
granular media filters, each with a 2 mgd capacity. In theory, the plant could have been
expanded in 2 mgd increments; however, in practice the cost of mobilizing-a contractor
and performing the necessary excavation and construction of a 2 mgd filter is
prohibitive. Therefore, the Marden WTP was expanded by an 8 mgd capacity increment
to allow economies of scale in the expansion.
By contrast, the membrane unit of production is a frame-mounted skid, which can be
economically -added to existing facilities to increase the production capacity in smaller
units. The proposed low-head pressurized membrane systems considered for Columbia
WTP will not require excavation or significant structural concrete construction for
addition of 1 to 2 mgd increments of expansion.
Production Flexibility
The membrane filtration option has more flexibility in production capacity than the
- granular media filtration option. This is because granular media filtration capacity is
limited by water quality consideration$; whereas, with membrane filtration the finished
water quality is _independent of production capacity. Membrane filtration allows the
opportunity to increase production by increasing production costs. For example
production capacity can be increased by increasing the frequency of chemical cleaning
without adversely affecting finished water quality. Granular media filtration does not
provide the same flexibility for increasing production capacity without jeopardizing
finished water quality.
, J
. j
PROCESS RECOMMENDATION
The recommended treatment process for the Columbia WTP is membrane filtration with
dissolved air flotation pretreatment and UV disinfection.
c j
, "
D:\Si's 0 Drive\EM-OO3.doc
United Water Idaho
Columbia Water Treatment Plant
Pilot-Scale Evaluation
THE EFFECT OF OAF PRETREATMENT
ON MEMBRANE PERFORMANCE
December 2001
UNITED WATER IDAHO
COLUMBIA WATER TREATMENT PLANT
PilOT-SCALE EVALUATION
THE EFFECT OF DAF PRETREATMENT
ON MEMBRANE PERFORMANCE
TABLE OF CONTENTS
Paae No.
INTRODUCTION.......................................................................................... 1
BACKGROUND ............................................................................................
METHODS ...................................................................................................
Description of Pilot System ...........................................................................
Water Quality Testing...................................................................................
RESUL TS.....................................................................................................
Water Quality Characterization .....................................................................
Hydraulic Performance of UF Membrane with DAF Pretreatment (August
2001) ............................................................................................................ 7
Hydraulic Performance of UF Membrane on Raw Boise River Water
(September 2001 )........................................... ..............................................
CONCLUSIONS......................................................................................... 12
Appendix A - Operational Parameters and Water Quality Data
Appendix B - Leopold DAF Pilot Plant Report
Table
Table 2
Table 3
Table A-
Table A-
Figure 1
Figure 2
Figure 3
December, 2001
LIST OF TABLES
Measured Water Quality Parameters............................................................
Water Quality Results Using DAF Pretreatment............................................
Water Quality Results Using Raw Boise River Water.................................... 6
DAF & UF Operational Parameters and Analytical Results .........................
Algae Sample Data 2001 ............................................................................
LIST OF FIGURES
Hydraulic Performance Data - Polymem UF Membrane Module -
Normalized Flux= 68 L/m ~ 20o
Boise River DAF Pretreatment Water........................................................... 8
Hydraulic Performance Data - Polymem UF Membrane Module -
Normalized Flux= 68 Llm ~ 20oC - Raw Boise River Water ..................
Hydraulic Performance Data - Polymem UF Membrane Module-
Raw Boise River Water - Normalized Flux= 55 L/m ~ 20o C ..................
Figure 4
Figure 5
Figure 6
Figure 7
December, 2001
Influence of Coagulation on Membrane Performance .................................
Hydraulic Performance Data - Polymem UF Membrane Module -
Normalized Flux= 68 L/m ~ 20o
Boise River DAF Pre-Treated Water........................................................... 14
Hydraulic Performance Data - Polymem UF Membrane Module-
Normalized Flux= 68 L/m ~ 20o
Raw Boise River Water............................................................................... 15
Summary of Pilot Testing data Chemical Cleaning Intervals .......................
THE EFFECT OF DAF PRETREATMENT ON MEMBRANE
PERFORMANCE
INTRODUCTION
United Water Idaho (UWI) is preparing to construct and operate a new water treatment facility
in the Columbia Bench area in southeast Boise that will use Boise River water as its source.
To evaluate potential treatment processes and determine future design and operational
criteria, a pilot-scale study has been conducted to determine the effectiveness of using
dissolved air flotation (OAF) pretreatment followed by membrane filtration. The main goal of
the pilot-study was to determine if OAF pretreatment would allow the membrane filters to be
operated at a higher flux compared to using raw Boise River water.
The pilot-study was conducted at the Marden Water Treatment Plant located in eastern Boise
which uses the Boise River as its raw water source. The pilot equipment consisted of a OAF
system provided by Leopold Water and Wastewater Products (Zelienople, Pennsylvania) and
a membrane pilot system provided by Carollo Engineers equipped with a Polymem
(Fourquevaux, France) ultrafiltration (UF) membrane filter. To evaluate the effectiveness of
using OAF pretreatment, the pilot-study was conducted in two phases. During the first phase
from August 4, 2001 through August 29, 2001 , the OAF was used as pretreatment for the
membrane pilot plant. During the second phase, from August 30, 2001 through September
28,2001 , raw Boise River water was used as the source water for the membrane pilot plant.
BACKGROUND
Membrane filtration is a pressure-driven separation process. Pressure is applied to one side
of a thin film (membrane) that provides a barrier to the transport of matter. In general
microfiltration (MF) and ultrafiltration (UF) are used to remove particulate matter in water, such
as inorganic and organic suspended solids and microbes. The type of matter retained by the
membrane surface is a function of membrane pore size and composition. Low-pressure
membranes, i.e. MF and UF, are used to remove microbial and particulate contaminants. MF
membranes are characterized by a typical pore size 0.05 J-lm to 0.J-lm. Through physicalsieving, MF membranes are capable of removing protozoan cysts (including Giardia and
Cryptosporidium) ranging in size from 1 to100 J-lm , and most bacteria which range in size from
1 to 100 J-lm. UF membranes have nominal pore sizes of less than 0.01 J-lm , which in
addition to protozoan and bacteria removal , allows UF membranes to effectively remove some
viruses. Pressure requirements are a function of the head loss through the membrane
(transmembrane pressure) and fouling due to the accumulation of matter on the surface of the
membrane, and within the pore structure. Operating pressures can be as high as 30 psi
depending on membrane manufacturer and source water quality. Low-pressure membrane
filtration provides a significant degree of disinfection by physically removing pathogens from
December, 2001
feed water. Consequently, chemical disinfectant demands are lowered , thus reducing the
potential for disinfection byproduct (DBP) formation.
Despite the potential benefits that low-pressure membrane filtration can provide in treating
surface water, capital costs can be higher than conventional granular media filtration and
disinfection technologies. High capital costs can be attributed to high fouling rates, low flux
high operating transmembrane (TMP) pressures, as well as high O&M costs due to energy
consumption and chemical use for periodic cleanings. Feed water quality significantly impacts
flux, fouling rates membrane replacement, and chemical cleaning intervals. In order to
increase flux rates and decrease fouling rates, pre-treatment systems are often used upstream
of a membrane. Several processes have been successfully implemented as pretreatment for
membrane filtration, including lime softening, GAC , direct coagulant addition , and in-line PAC
addition. More recently, dissolved air flotation (OAF) has been suggested as an appropriate
pre-treatment for membrane filtration.
Reduction of suspended and dissolved particulate matter, including total organic carbon
(TOC), from the membrane feed water was a main goal of the OAF pretreatment process for
the Boise River pilot study. The typical OAF process consists of coagulant injection , rapid mix
and flocculation, and removal of the floc from the water through a floatation process. The
floatation step is accomplished by injecting water super-saturated with air at high pressure into
a reactor with a free water surface. The de-pressurization of the water causes micro bubbles
to form and rise to the surface of the tank. Under the right conditions, these air bubbles attach
to the floc particles and float them to the surface of the reactor. The resulting floating sludge
blanket is then skimmed into a trough for disposal. The process is well suited for the removal
of algae from a surface water supply and is capable of removing suspended solids, TOC and
reducing turbidity.
The goal of the study was to evaluate the influence of OAF pretreatment on UF membrane
hydraulic performance on Boise River surface water. For a variety of reasons, including the
ability of OAF to remove algae from raw water, OAF pretreatment was selected for this study.
The experimental approach was to operate the UF membrane pilot system with OAF effluent
as its feed water at an aggressive membrane flux rate (68 Llm2 -
:!:
~ 20oC in dead-end
mode, Le., without recirculation) for a one month period (August 2001). The objective was to
determine if membrane performance could be improved over the previously obtained
membrane performance when operating without OAF pretreatment, as given below:
Flux: 40 L/m ~ 20oC (24 gpd/ft2
Mode: dead end
Flow rate: 21 gpm ~ 20o
Backwash frequency: 60 min.
Percent recovery: 90%
December, 2001
The above operating conditions were obtained with a Polymem UF120 S Ultrafiltration
membrane module optimized on raw Boise River water during pilot testing conducted between
October, 2000 and March, 2001. The aggressive membrane flux rate of 68 Um h :t (g120o
with the OAF pretreatment was selected based on this prior pilot work.
The conditions selected for the first phase of the pilot-study (August, 2001) using OAF
pretreatment were also used for the second phase of the pilot-study (September, 2001) in
which raw Boise River water would be used as the membrane feed water. A third phase of the
study was also performed using raw Boise River water at a flux rate less than 68 L/m , but
greater than the previously determined optimum flux rate of (40 Um h). The methods and
materials used in the study are described in the following section.
METHODS
Description of Pilot System
The Leopold OAF pilot unit consisted of a two-stage flocculation basin followed by dissolved
air flotation. Ferric chloride (FeCI3) coagulant and at times, hydrochloric or sulfuric acid were
added for pH adjustment to the OAF influent (raw Boise River water) prior to the water entering
the flocculation basin. No polymers (coagulant aids) were used in the study as their use
generally has a negative impact on OAF and membrane performance.
To determine the impact of varying the OAF surface loading rates (SLR) on membrane
performance, two conditions were evaluated. The initial conditions (8/4/2001 through
8/19/2001) consisted of a SLR of 4 gpm/sq ft (36 gpm total), an optimal flocculation energy
(Gt) of 60 000 sec , flocculation time of 20 minutes, and a recycle flow rate of 12.5%.
Subsequent conditions (8/19/2001 through 8/29/2001), consisted of a SLR of 8 gpm/sq ft (72
gpm total), an optimal flocculation energy (Gt) of 30 000 to 40 000 sec , flocculation time of
10 minutes, and a recycle flow rate of 6.5%.
The membrane pilot plant was equipped with a 500 micron prescreening strainer followed by
membrane filtration. A Polymem UF module that has an outside-to-inside hollow fiber
configuration utilizing a combined pneumatic and hydraulic backwash was used during the
study. The specifications for the UF module are as follows:
Membrane Surface Area:
Material:
Nominal Pore Size:
Configuration:
114 m
Polysulfone
01 J.lm
Hollow Fiber Out/In Dead End Filtration
During the first phase of the study using OAF pretreatment, the membrane pilot plant was
operated at a flux of 68 L/m h with 93% recovery. During the second and third phases of the
study without OAF pretreatment, the membrane pilot plant was operated at a flux of 68 Um
December, 2001
and 55 Llm , respectively, with 90% recovery. These two final phases of the study allowed
for evaluation of the impact of OAF pretreatment on membrane performance.
Water Quality Testing
The water quality tests performed during the first phase of the pilot-study are presented in
Table 1. The turbidity and pH of the OAF influent and effluent were recorded continuously by
on-line instrumentation on the OAF pilot unit. Raw water temperature and dissolved oxygen
levels were also recorded continuously by on-line instrumentation on the OAF unit. Total iron
and dissolved iron in the OAF effluent were measured on-site daily by the OAF operator using
colorimetric method and Hach equipment. Raw water and OAF effluent samples were
collected three times during the pilot-study and analyzed for total organic carbon (TOC),
dissolved organic carbon (DOC), UV absorbance at 254 nm, and algae enumeration and
identification by MWH Laboratories. DOC, UV 254 absorbance, and a simulated distribution
system (SDS) disinfection byproduct (DBP) analyses were also conducted at the same times
on the membrane permeate water at MWH Laboratories. Total trihalomethanes (TTHM) and
haloacetic acids (HAA5) were measured to evaluate the 48-hour SDS DBP formation.
Table 1 Measured Water Quality Parameters
Columbia Water Treatment Plant Pilot Study
United Water Idaho
On-Line Measurements On-Site Analysis Off-Site Analysis
Turbidity Dissolved iron TOC
Total iron DOC
Dissolved oxygen Manganese UV 254 Absorbance
Temperature TTHM
HAA5
During the second and third phases of the pilot-study (without OAF pretreatment), turbidity and
pH of the membrane influent (raw Boise River water) and permeate were measured. These
samples were collected daily by the staff of the Marden Water Treatment Plant and analyzed
on-site at the Marden laboratory. In addition, the DOC of the membrane influent and permeate
was analyzed once along with SDS DBP analyses on the membrane permeate at the MWH
laboratory.
December, 2001
RESULTS
Water Quality Characterization
Throughout the first phase of the pilot-study, the ferric chloride (FeCI3) and acid doses were
varied to optimize the OAF effluent turbidity. Initially (8/4/2001 - 8/13/2001), 12 mg/L of FeCI3
and no acid were added to the OAF feed water. From 8/14/2001 to 8/17/01 , hydrochloric acid
was added to suppress the pH to around 6.1 to optimize coagulation so that a lower FeCI3
dose of 8 mg/L could be used without effecting OAF effluent turbidity. From 8/18/2001 until
8/29/2001 (the end of the OAF study period), sulfuric acid was substituted for hydrochloric acid
for pH suppression, and the FeCI3 dose varied from 8 to 12 mg/L. The turbidity of the OAF
effluent averaged about 0.5 NTU and ranged between 0.45 to 0.6 NTU. With acid addition
(using either hydrochloric or sulfuric acid) the pH of the OAF effluent remained around 6.
4. 1. 1 Raw Water and Membrane Feed Water
The FeCI3 and acid dose added to the OAF feed water was optimized based on turbidity.
TOC , DOC UV 254 absorbance , and algae removals provided by the OAF were also
observed. The results of the water quality analyses conducted during the first phase of the
pilot-study are presented in Table 2.
Table 2 Water Quality Results Using OAF Pretreatment
Columbia Water Treatment Plant Pilot Study
United Water Idaho
8/9/2001 8/18/2001 8/29/2001
OAF SLR (gpm/ft2
Flocculation Times (min)
FeCI3 Dose (mg/L)
SO4 Acid Dose (mg/L)19.
Raw Water:
TOC (mg/L)1.42
DOC (mg/L)1.47 1.42 1.41UV 254 Abs.(1/cm)039 039 037Algae(#/mL)286 139 (8/28/01)Dissolved Iron (mg/L)
OAF Effluent:
TOC (mg/L)
DOC (mg/L)
UV 254 Abs.(1/cm)018 017Algae(#/mL)
Dissolved Iron (mg/L)~0.~0.~0.
Membrane Permeate:
DOC (mg/L)
UV 254 Abs.(1Icm)015
SDS TTHM (J-lg/L)
SDS HAA5 (J.lg/L)
December, 2001
At a SLR of 4 gpm/sq ft, a FeCI3 dose of 12 mg/L and no acid addition , little removal of TOC
and DOC (50/0 to 16%) by the OAF was observed. At a SLR of 4 gpm/sq ft, a FeCI3 dose of 8
mg/L and a sulfuric acid dose of 19.2 mg/L, a significantly higher removal of TOC and DOC
(30% and 39% respectively) was observed. With a SLR of 8 gpm/sq ft, a FeCb dose of
mg/L and a sulfuric acid dose of 16 mg/L, some removal of TOC and DOC (18% and 35%
respectively) was observed. The UV 254 absorbance removal averaged 52% for all of the
OAF operational conditions tested.
Algae enumeration was conducted on raw water (as part of United Water s weekly sampling
program on the Boise River) and on OAF effluent on 8/9/01 and 8/28/01. Algae removals of
85% and 73% were observed on 8/9/01 and 8/28/01 , respectively. An additional OAF effluent
algae enumeration sample was taken on 8/18/01 to provide additional data on membrane feed
water. The three OAF effluent algae enumerations (44, 31 , 28 #/ml) indicate that significant
algae reduction occurred throughout the first phase of the study.
Membrane Permeate Water
The data provided in Table 2 for the membrane permeate water indicates that the membrane
filtration process removed little or no dissolved organic matter based on the DOC and UV 254
absorbance analyses. The SDS DBP analyses conducted resulted in the average formation of
TTHMs at approximately 10 Jlg/L and HAA5 at approximately 36 Jlg/L resulting from theaddition of chlorine (2.0 mgll of NaOCI) to the membrane permeate water. The current
maximum contaminant levels (MCL) for these two regulated chlorinated DBPs are 80
Jl9/L for
TTHMs and 60 Jlg/L for HAA5, which are greater than the levels measured during the analysis.
The water quality analysis results from the second phase of the pilot-study are presented in
Table 3.
Table 3 Water Quality Results Using Raw Boise River Water
Columbia Water Treatment Plant Pilot Study
United Water Idaho
Boise River
Water Membrane Permeate
Date
9/25/01
* From UWI weekl
Only one membrane permeate sample was analyzed during this second phase of the study
while the membrane unit ran on raw Boise River water. The sample taken on 9/25/01
indicated that DOC reduction was negligible through the membrane unit (1.40 mg/L influent
and effluent samples). The higher permeate DOC and UV 254 absorbance levels led to higher
SDS DBP formation for both TTHM's (29 Jlg/L) and HAA5 (54 Jlg/L) when compared to the
December, 2001
- --
fJ=~
_-
i~t
__--,,-
~.. ~-~ ~n"~
.~._- -"'-.---- ~-.."..,----_..._.
_o~
.~--~"~~'--~- .~~...-,.._.~.._-.~. ..._---_-~---.~, ~"'-'_~-"~.-~",., '-. "'.--.--.-.-.""~,~. .,~---_. -- --..- ~~- -.-. .. -.-.. --.. ~-.._" - ----~-~--= -- '" "" ~--" -.. ~- .,,-,..~... ~ -- "'--.=-~.....
-. Oh.
-. ,-" '.. - -~" - -"" --, ~---~ ---." -.,..., "'-~.- -"-~ "-.-.~- ~ ~ "",_. -=... ~ ---"'. _.~,---_.~" ~---- -~_- ---~.~~--_......, ._~=---~----".,-=_..~~.~._..._._.__.,~_.. _..- _....~.,,-.._- .-~ ----'-. ..~~~~.~._.~---_..._=..~.--~~- -~ -'- . -~_.--......_~~=-,.~,,--_.".~ '"=~".
-""'_~C"-_.
,.,,. .
,"_C.-
~~.,,-.-.".,."..~ ....~~~'".., ~~,. ""~.-..--,
("
)
...
,
J\
.
)
Ci)c..CD

..
.

Co
)
~
0
~
-
;
-

1
5
...
a.
.

.
a
CD

E
s:
:

C
D
ca

c
.
.
.c

E
CD

1
0
CD

E
S
s::
..
.
= =

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=
=

=

=

=

=

=

=

=

=

=

~=
~
=
~
=
~

=
=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=
-
-

-

-

-

-

-

-

-

-

~~
r

e
m
a
~
-

= =

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

~I
T
n
~
~
d

f
~
x

~
W
~
C

=

= =

=

=

=

=

=

=

=

=

=

=

=

=
=
=

=

=

=

=

=

=

=

=

=

cr
I
e
=
a
t
n
!
e
D
(
~
s
J
~
J
i
Q
n
J

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-
Ti
m
e

(
D
a
y
s
)
25
0
20
0

~

z
en

"C

..
.
CD

3
~.

e!
.
.
"'1
1

c
.
.
"'1
1
)(

~
..
.

C
3..
.
10
0

g

2
:
(8
)
(8
)

~
~
~
~
n
Fi
g
u
r
e

1
BO
I
S
E

R
I
V
E
R

D
A
F

P
R
E
T
R
E
A
T
M
E
N
T

W
A
T
E
R
Hy
d
r
a
u
l
i
c

P
e
r
f
o
r
m
a
n
c
e

D
a
t
a
.

P
o
l
y
m
e
m

UF

1
2
0
S
2

M
e
m
b
r
a
n
e

M
o
d
u
l
e
No
r
m
a
l
i
z
e
d

F
l
u
x

=

6
8

U
m
h
~

2
0
o
(+
1
.

5
U
m
CO
L
U
M
B
I
A

W
T
P
Hydraulic Performance of UF Membrane on Raw Boise River Water
(September 2001)
Following the phase 1 testing with DAF pretreatment, there were 2 runs performed on the raw
Boise River water. Termination criteria for membrane operation for phase 2 were consistent
with phase 1 testing with DAF pretreatment. The operational parameters of normalized flux
backwash interval, and recovery during the first run of phase 2 were identical to the DAF
pretreatment run. This allowed direct comparison of membrane fouling rates as a method of
evaluating the impact of DAF pretreatment on the membrane operation. The operational data
obtained during the first raw water membrane run is provided in Figure 2. The TMP increased
rapidly from an initial value of 6.5 psi to the terminal limit of 15 psi after about 0.48 days (11.
hours) of operation and the run was terminated. Due to the short run time associated with the
first run, a second run on the raw Boise River water was conducted with the following
membrane pilot plant conditions:
Flux: 55 l/m ~ 20oC (32 gpd/ft2 ~ 200
Flow rate: 27 gpm
Backwash frequency: 50 min
Percent recovery: 90%
The data provided from the second run is presented in Figure 3.
The TMP rose from an initial value of 6 psi to the terminal TMP of 15 psi after about 1.5 days
(36 hours) of operation and the run was terminated.
The potential for enhanced performance depends on the nature of the foulants in the feed
water, and their potential for interaction with the membrane. Based on the observations made
during the study, there are several possible mechanisms responsible for the increased flux
rate observed during Phase 1 testing. These include:
Reduction in algae counts, which have been demonstrated to cause high fouling rates in
membrane filtration
Reducing foulant penetration into membrane pores
Conditioning of the layer of materials deposited in the membrane , or
Improving particle transport characteristics.
The last three mechanisms are also typical of coagulant addition alone ahead of a membrane. It was visually apparant during backwashing of the membranes that iron floc had
carried over from the DAF unit, suggesting that reduction in algae counts alone was not solely
responsible for the improved performance. Further evidence that iron floc aided in theimproved hydraulic performance is provided in the following discussion.
December, 2001
:!
!
("
)
I'.
)
=-
-
-Q)

(.
)
UJ

UJ

Q
)
...
D.
.

.
a
Q)

E
s:
:

Q
)
co

c
.
.a

E
Q)

1
0
Q)

l
-
E
~
s:
:
...
Tr
a
n
s
m
e
m
b
r
a
n
e

P
r
e
s
s
u
r
e
Te
m
p
e
r
a
t
u
r
e
I
:

:

:

:

:

:

25
0
'T
e
r
m
i
n
a
l

T
M
P
Sp
e
c
i
f
i
c

F
l
u
x

~
2
0

C
No
r
m
a
l
i
z
e
d

F
l
u
x

~
2
0

C
--
-
-
-
-
-
-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-
-
-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-=

-

~-

-

-=

-

~-

-

~

-

-
--

-

~

-

-
--

-
--
r
-

1
5
0
-
-

=

=

=
=
=

=

=

=

=~
r
~
~
~
~
~
~
~

=
=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=

=
=

=

=

=

=
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

--

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

--

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
0.
4
Ti
m
e

(
D
a
y
s
)
1.
4
Fi
g
u
r
e

2
RA
W

B
O
I
S
E

R
I
V
E
R

WA
T
E
R
Hy
d
r
a
u
l
i
c

P
e
r
f
o
r
m
a
n
c
e

D
a
t
a

-

P
o
l
y
m
e
m

U
F

1
2
0
5
2

M
e
m
b
r
a
n
e

M
o
d
u
l
e
No
r
m
a
l
i
z
e
d

F
l
u
x

=

6
8

U
m
h
~

2
0
o
C
(
+
/
-

5

U
m
Re
c
o
v
e
r
y

=

9
3
%
CO
L
U
M
B
I
A

W
T
P
20
0

0
.,

z
en

0
"C

.
,
CD

3
::
:
;
;

=-
-
-.

N
"'1
1

C
.
"'1
1
)(

C
..
.
.

C
:s
-

3
...
.
00

a=

:s
-
O)

..
:
:
!
.

(
8
)
(8
)

N
N
~
(')
50

(')
(')I\
.
)
==
-
c.
.~
0
U)

U)

C
P

..
.

5
-
..
.

~
D.
.

CP

e
c.
.
.c

E
~
~

U)

0
..
.
Tr
a
n
s
m
e
m
b
r
a
n
e

P
r
e
s
s
u
r
e
Te
m
p
e
r
a
t
u
r
e
-
-

'T
e
r
m
i
n
a
l

T
M
P
Sp
e
c
i
f
i
c

F
l
u
x

~
2
0

C
No
r
m
a
l
i
z
e
d

F
l
u
x

~
2
0

C
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-
--
-
-
-
-
-
-
-
-
-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
--
-
-
-
-
-
-
-
-
-
-
-
-
-
-
--
-
-
-
-
-
-
-
-
-
--
-
-
-
-
-
-
-
-
-
-
-
-
-
-
--
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-

-

-

-

-

=,
=

= =

=

=

=

=

=

=

=

=

=

=
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
1=
e
r
m
i
R
a
I

T
M
P

g
f
-

-
-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

t~
~
r
R
~
c
~
d

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-
-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-
-

-

-

-
0.
4
Ti
m
e

(
D
a
y
s
)
1.
4
Fi
g
u
r
e

3
RA
W

B
O
I
S
E

R
I
V
E
R

WA
T
E
R
Hy
d
r
a
u
l
i
c

P
e
r
f
o
r
m
a
n
c
e

D
a
t
a
.

P
o
l
y
m
e
m

U
F

12
0
S
2

M
e
m
b
r
a
n
e

M
o
d
u
l
e
No
r
m
a
l
i
z
e
d

F
l
u
x

=

5
5

U
m
h
~

2
0
o
C
(
+
/
.
5
U
m
Re
c
o
v
e
r
y

=

9
0
%
CO
L
U
M
B
I
A

W
T
P
25
0
20
0
..
.

z
en

"C

..
.
~
3
-.

-
-
o'

N
"
"'
1
1

C
D
c.
.
"'1
1
)(

-
-
c
=l

C
:!
r

=
l
C"

:
r
DJ

..
:
:
!
.

(
9
)
(9
)

N
N
Q
("
)
("
)
15
0
10
0
Figure 4 is a graphic representation of the influence of coagulation on membrane
performance. Although detailed analysis of the mechanisms responsible for the observed
increased flux rate was not conducted, it is reasonable to describe the enhanced performance
in terms of this model. Evidence of enhanced backtransport and reduced pore penetration can
be seen in Figure 5 (with OAF pretreatment). This figure shows the TMP trend over three
filtration cycles. During day two of the OAF test, there was a sharp rise in TMP over the
course of the production cycle (from about 5.5 to between 6.2 and 6.8 psi), indicating a
buildup of material on the membrane surface. However, following the backwash cycle, the
TMP returned to about 5 psi. This TMP recovery indicates that the membrane was reversibly
fouled and that there was near complete removal (backtransport) of solids during
backwashing.
Over the course of the OAF/membrane testing, some fouling did occur; however, compared to
the raw water testing (without OAF pretreatment), the fouling rate was significantly lower. A
three hour TMP trend for the first raw water condition is presented in Figure 6. Over this
interval there is a noticeable increase in TMP at each data point following a backwash , as well
as a rapid overall fouling rate. This rapid overall fouling led to an 11.5 hour chemical cleaning
interval.
0 CONCLUSIONS
Based on the results of this study, it can be concluded that the OAF pretreatment dramatically
increased the treatment capacity of the UF membrane filtration module used for this study.
Under the conditions tested using OAF pretreatment, the membrane TMP remained below
psi for the 25 day run. The data suggest that possibly higher flux rates (:::- 68 L/m ~ 200
could be achieved with OAF pretreated water even at the high SLR of 8 gpm/sq ft, while
maintaining the 30 day chemical cleaning interval. During the second phase of the pilot-study,
decreased membrane performance using raw Boise River water without pretreatment was
clearly shown. At a flux of 68 L/m h ~ 20oC, terminal TMP was reached in 0.5 days (11.
hours), and at a flux of 55 L/m h ~ 20oC terminal TMP was reached in 1.5 days (36 hours).
(Figure 7).
December, 2001
("
)
""
"
I'
\
)
Wi
t
h
o
u
t
Co
a
g
u
l
a
t
i
o
n
Wi
t
h
Co
a
g
u
l
a
t
i
o
n
(J
.
)
In
c
r
e
a
s
e
d

P
o
r
e

P
e
n
e
t
r
a
t
i
o
n
..
,
G
8
Re
d
u
c
e
d

P
o
r
e

P
e
n
e
t
r
a
t
i
o
n
Le
s
s

P
o
r
o
u
s

C
a
k
e
~~
~
~

~

~
~
~
~
~
~

~

~
~
~

~
~

~
..
.
Mo
r
e

P
o
r
o
u
s

C
a
k
e
Re
d
u
c
e
d

B
a
c
k
t
r
a
n
s
p
o
r
t
~
~

,&
8
..
-
-
.
En
h
a
n
c
e
d

B
a
c
k
t
r
a
n
s
p
o
r
t
Fi
g
u
r
e

4
IN
F
L
U
E
N
C
E

O
F

C
O
A
G
U
L
A
T
I
O
N

O
N
ME
M
B
R
A
N
E

P
E
R
F
O
R
M
A
N
C
E
CO
L
U
M
B
I
A

W
T
P
("
)
""
"
'
"
""
"
'
"
.j:
:
:
.
.
10
.
0
T
r
a
n
s
m
e
m
b
r
a
n
e

Pr
e
s
s
u
r
e
..
-
-

Ba
c
k
w
a
s
h
:"
"
I
Pr
o
d
u
c
t
i
o
n

C
y
c
l
e
TM
P

P
r
i
o
r

t
o

B
a
c
k
w
a
s
h

.0
'
~
.."
,
.
.
.
"
-
0
"
...
po
.
..

.
.

0"

.
.

'o
J
'o
J

..
G
-

~
0.

~
.
-
..
(
5
...
-e
5
TM
P

P
o
s
t

B
a
c
k
w
a
s
h

==
-
C1
)

..
.

u
~
0
en

en

C1
)
C1
)

.
.
.
Q:

~

7
.
C1
)

.
.
.
s:
:

C
1
)
C'
O

c
.
.a

E
C1
)
E
~

6
.
s:
:C'O
..
.
2.
2
1
2.
2
2
Ti
m
e

(
D
a
y
s
)
Fi
g
u
r
e

5
BO
I
S
E

R
I
V
E
R

D
A
F

P
R
E
-
TR
E
A
T
E
D

W
A
T
E
R
Hy
d
r
a
u
l
i
c

P
e
r
f
o
r
m
a
n
c
e

D
a
t
a

-

P
o
l
y
m
e
m

U
F

12
0
S
2

M
e
m
b
r
a
n
e

M
o
d
u
l
e
No
r
m
a
l
i
z
e
d

F
l
u
x

=

6
8

U
m
h
(
g
1
2
0
o
C
(
+
/
-
5
U
m
CO
L
U
M
B
I
A

W
T
P
(')I'V
...
.
i
o
.
14
.
:=
-
13
.
G)

..
.

U
::
s

0
U)

U)

G
)
G)

.
.
.
..
.

:
:
s
0.
.

G)

E
c:
:

G
)
12
.
~
c
.
15

E
G)

I
-
..
.
U)

0
c:
:
...
11
.
10
.
0
T
r
a
n
s
m
e
m
b
r
a
n
e

Pr
e
s
s
u
r
e
Pr
o
d
u
c
t
i
o
n

C
y
c
l
e
0'"
/'

T
M
P

Pr
i
o
r

t
o

B
a
c
k
w
a
s
h
..
...
.
.
.
.
,-
.
.
.
""
"
...
0.
/
...

...
TM
P

P
o
s
t

Ba
c
k
w
a
s
~
TM
P

P
o
s
t

B
a
c
k
w
a
s
h
,.
.
.
.
.

"
'7
\
.
Ti
m
e

(
D
a
y
s
)
...
.
i
o
.
CJ
1
Fi
g
u
r
e

6
RA
W

B
O
I
S
E

R
I
V
E
R

W
A
T
E
R
Hy
d
r
a
u
l
i
c

P
e
r
f
o
r
m
a
n
c
e

D
a
t
a

-

P
o
l
y
m
e
m

U
F

12
0
S
2

M
e
m
b
r
a
n
e

M
o
d
u
l
e
No
r
m
a
l
i
z
e
d

F
l
u
x

=

6
8

L
/
m
h
~

2
0
o
C
(
+
/
-
5
L
/
m
Re
c
o
v
e
r
y

=

9
3
%
CO
L
U
M
B
I
A

W
T
P
~3
0

d
a
y
s

(
e
s
t
.
)
s::
:
s:
:
:
.!
:
!
Co
)
.s
:
:
:
Co
)
5
-
5
d
a
y
s
DA
F

P
r
e
t
r
e
a
t
m
e
n
t
Ra
w

B
o
i
s
e

R
i
v
e
r

W
a
t
e
r
Fl
u
x

=

6
8

L
/
m
2 -
Fl
u
x

=

6
8

L
/
m
Re
c
o
v
e
r
y

=

9
3
%
Re
c
o
v
e
r
y

=

9
3
%
Au
g
u
s
t

2
0
0
1
Se
p
t
e
m
b
e
r

2
0
0
1
('
)C"
'
I\
.
)
-:
:
1
.
5
d
a
y
s
Ra
w

B
o
i
s
e

R
i
v
e
r

W
a
t
e
r
Fl
u
x

=

5
5

L
/
m
Re
c
o
v
e
r
y

=

9
0
%
Se
p
t
e
m
b
e
r

2
0
0
1
~3
0

d
a
y
s
Ra
w

B
o
i
s
e

R
i
v
e
r

W
a
t
e
r
Fl
u
x

=

4
0

L
/
m
Re
c
o
v
e
r
y

=

9
0
%
De
c
e
m
b
e
r

2
0
0
0
Fi
g
u
r
e

7
SU
M
M
A
R
Y

O
F

P
I
L
O
T

T
E
S
T
I
N
G

D
A
T
A
CH
E
M
I
C
A
L

C
L
E
A
N
I
N
G

I
N
T
E
R
V
A
L
S
CO
L
U
M
B
I
A

W
T
P
tr
j
'1f\" \i'n::r, t,! \i'\1'"Ji"( (;"TI1:~"1\ "T\' j't,r'\f ll.. J:d\. Ql w ~ l"1"
.)
;t ~. J. ,1;'
PRODUG1~S
_.~
MARDEN STREET WATER
TREA TMENT PLANT
Boise, ID
LEOPOLD DA.
PILOT PLANT REPORT
~ummer Study
August, 2001
Author: MattScholIlaker
Operator:Matt Schomaker
United Water Idaho
Boise, Idaho
, I
Leopold Dissolved Air Flotation PilotReport
Boise River Water at the Marden Ste W ater Treatment Plant
IJQise,
August 2001
Author: Matt. Schomaker
Operator: MattScbomaker.
Client:U nited \Vater, Idaho
Engineer: Moptgomery Watson Barza
BoiSe,.
Jl~viewed By:
Bill Marticorena
ProductManager- Clairit1cation.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise. Idaho
Contents
Executi yeS ummary .
. . . . . . . . . . . . . . . . . . . . . .
0 . 0 0 . . 0 . . . 0 . . . 0 . . "0 . . . . . . . '
. . . .. . . .
... 5
. . . ~.. . . . . . . . . . . .
0 . . . . . 0 . . . . ~ ~ o. . . . . . 0 0 . . ..
. . . . . . . . . . . . .
' o. . . 0 o. ..
. 0
... . . .
0 .. . .
. . . . . . . . . . . . .
0 . . . . . . . . . '. 0 . . . 0 . . . . 0 . . . 0 . . . . .
. . . ... . .
Introduction. . .. . 0
Qbj ecti ves 00 . . .. .0 ,
4. Metl1,odo logy. . . .. . .. o.
.. . .
' . 0 . . . . . . . . 0 . 0 o. .
. . ... . ~ . . ~ . .. . . . .
. . . . . . . . . . . o. . . . . . . . ..
Results..... .
. . ..
0 . . . . .
.. . . . . . . . . . . . . .
~ . . . . . 0 '
' . .. .. .. .. .. . '' . ... . . . . .. . . . . .. . .. ... .
awWatetQu~ti~y....... '
" "'" .... ". '" ~."'" ......... ,. ". .., ..,. ..
1.2.
1.3.
1.3.1. ApparentC()lor,..., 0'."
. "...... .:... ..., .~.' ;...,..... ......... ,. ..,. .
2. Tru.(! .Colo,..
. :, '' , . . . '" ,. . . . . . . . . ,. .. . . . . . . . .. ". . '. . . . , . , ,. . . , . . . . . . . .
,'" 11
1.4. RaW WatltrlrOTl Concentration.,. . o. .
~. ....;. .. . . .:... '.... p,.' . .:,. ,.. . . . . '
;' 13
.4.1. 1;i!tallton..
,..
5.1,2. Dissolved/ron.
;.~.. .
n.. .
,..... ... ".:, . ,. .,., ........,... .. ..~.... .:.. ... , ..
5, Raw Water Aluminum............ o. . .H. . . .'" o..' i .. .., .. to
.. "' . ,.... .. .
,. ,... .., 15
5.1 ;6, Ro:w Wilt#r Matlganes(!......
, "
," ..n, ...,., .
.... ... "
"'...o... 15
. .
1.7. RawWqtefLJ.issolvi!dOXygel1.,.. i.. ...i...;. ..i . ;..~..~;;. ;.,..,.
:' ....
,..':',..,.. 16. 5.1~& RalV 'd"er1'~1tlPeiature.. . 0" '0' .. ,'"
... .. ..
oo ,.,.
~... ....., ., .. ..".~ ... ...
:Pj$S9IY~C:1.;\i'('PJOJl!tf9r.,.
;~.., ," ,..~.. . .,. . ""~,. '
, '0"
'" """.",. ,. ';.' . ,
5..1. DAF' TZl.rb.iiJity..............
..,. .,." ........ ,........ .......,..,....,..... .... .....
2;~. P4Fp'fl),...,... 'f' ... " ..
...., "., ; .,'" ,'" '" ',." ". .. ,:.~. "
0 ..
...,.
f' 1&
2.3.. qqll.gI1Jtlti()fJ 'oo ..... ,..., "'0.. ~... ....
,...,... ,............
"'oo.oo. ....
.........
4. F!l!cfClflil#()lJ.Al1ergy
..~.~.. . "
o',.,..,.",.
~...,."..,..
~22
C()lorRiFm(Jvlll...
~.... ....~. . ,;.... . .' . . ~. ....~.. . . . .. .,. . .... . .. .. ... .'" . .+, ... . .... ., .?-.
1. .A,p/xJ.re#((jQIQr......,....... .'H"
.' ...', "'" .,..,......." '" . .. ....
.5.~~5.2. Tf'ue.(JoIQt.
. .. .. ""....
~.........,...o.o',..
6~ It()n(J(J"J,(::entratioll..
;~.... ,.... "
0' '
..'" '
0' "'.o""
" .
.. '0' ...,.. .... 'o'....~25
2./1. Milhg(J,lle~e.CdllC~~trti#()#, .f"
' .. ...". .... ....:.. '" '
' 'o.~'"
~ "" .
2.72l8.DA.FSluclg~.,.
............ ......... """" ....... ""'" .... ,. ..,. .... .... ....
~28
1. :SotldsConfeni..,.,...., .." u
" .... .... . ... . ... .. ... , ... ..., . . , . " .. , . . ,. ...
RawWaleiCQlor..........,
... ~ ,..,. '
'...,... ." ... 0"'.' '0. .... .....
..:..,.
... 11
Raw
Raw
Water Turbidity.,...
, ., . . . ... .. .. ... ... . .. . ... " " . .
w.. .
.... . "." . .... . .
Water rill...
" ''.' ",' ''" '.'.'" ........ ... ... "'" "'" ..... ..., .... ."" '"."" .... ...:......... ... .....,... ''..' ...... .
Sl~4gePr()tlui.;ti9ri'V(Jlume....
. ... . '
..... 0
... ". ~.. .~. "." . ~.. ...~.. . . .
~78
Conclusi()n.....
....~.... ..... ."
Montgomery Watson Hafia. Com pa6yCol1fi denti a
United Water Idaho
Boise. Idaho
List of Figures
Figure 1. Raw Water Turbidity. ..
. .. . .. . . ..... . . . . .. . ... . .... . . , . .. . . , . , . . .
... . 9
Figure 2. Raw Water pH
.... -"
.. 'oo... . . .
. . .. . ... . .. ...... "' . . '. .. ... ... . ..
.... 10
Figure 3 . Raw Water Daily pH Cycle.
, .. , .. , .. .. . .. .. .., .. . .. - , . .. .. .. .. .
. .... 11
Figure 4. Raw Water Apparent Color Residuals...............
-........ -.. "
, 12FigureS. Raw Water True ColotResiduals....... .....
,....,.........."
..,.. 12
Figure 6. Raw Water Total ItonConcentration. .
. .. . .. . . - . , . . , . . . . . . . . . .. ...
Figure? Ra.WWatefDissolvedIiOhC6ncellttation........................ .
Figure 8. Raw Water Manganeseeoncentration..... .. . .o. ';0.
... . . ... . . . . . ...
Figure 9. Raw Water QissQlyeq Oxygen Concentrati()n~,....... .."
...,. ...
Figure lO.Raw WatyrTefi:1petature,................ 0.
" ~..'" ..... ~."..,.... .
Figut~ 11. R.aWfi;\1JdJ)Af'tll,r1jidity~...
..,.. ,., "
" 0,"
, ... ,''.'.' "'.'
p" 18
Figl1r~ l:4.PAFEft1l.len.tP:E-I "
, ..... ... ... . ..... . .. . ,," ..... ."" ... "."."" "
." 19Figure 13. FenicQpfimizationat4gpmlft without pH Adjustment....
...
Figure 14. TurbidityOptitnizationwith pH AdjllstInent and FeCI 3 Dose
pf 12I11g11.
.... "...
'i"o'
,,"" "" . . .. .... . , .
.. H' ... . .. io. '"t.." .
'..' ...
21 '
Figure 15. TurbiclityQptinriia!jOI1with FeChJ)o$ecmdptIAdj1.1.stJri~nt..
Figllr~ 16. .1u.tlijdity():pti$i~~ti9i1witb.. P~.Gh .PQs~~clpH1\g.j(i,~tmept..2AFigure11.FloC9111~tiQli.Enefgyat. 4gp111Ift
. ....... -.. .... ... ...........
... u
' ~
Figure. 18. FlocctilatioI1EnergY:4J8.gptrilft2
.. . ~.. ."... ." .. .,: . ..,:...~... ...... .~.
FiglIre 19~:Ra w~(LDAFApp~ent(3()lQt 'f" .....
~....~.. '~'.~'~. .....
t~... Figure 20. RawatldI.)A.F.triJe(39161".
~..,,~ ~'.~.."....':. ..'.""~.' .,;,....
~ .~25
Figuf~21. RawaildDAFl)Jtallrpn..... ...,...
... ""~".."'" .'
'.I In "'.." .... f... 26.
Figul'~ '22. Ra\Vai!dDAF,f)i$$9!Y~d.lj'()J1.
. ',,' ..... ...,....,;; ..~...............~.
2(iFigure 23. Raw.anclpA.F1\1~ltl.ganes~~'f'~.'.
".,
......"..f... '."
,"'" .." ....... ... ..
MontgQm~ry Watson Harza Co mpany COtlt)9,Yl1tia 1
United Water Idaho
Boise. Idaho
List of Tables
Table 1. DAF Sludge Percent Dry Solids...................................
Table 2. OAF Sludge Production Volumes.................................
MontgomeryWatsdJ1.J-Iatza Company Confidential
United Water Idaho
Boise. Idaho
1. Executive Summary
TheF.Leopold Co. Inc. conducted a dissolved air flotation (DAF) pilot study in the
summer of 2001 at the Marden StWater Tre~tment Plant in Boise, Idaho. The study ran
from 1 August 2001 to 31 August 2001. As part of this study, the team of Montgomery
Watson lIarza (MWH)andCarollo Engineers conducted an ultrafiltration membrane
pilot study on the DAF effluent.
TheDAF pilot system was set-upto run~t 4gpm/ft2 forthefirsthalfofthe study with all
of the eftluent Water going directly to the membrane pilot unit~For the second halfofthestudythe.DAF unit was run at 8 gPmlft2 With a portion oftbe :PAF~ffh.l~nt going to the
membran~ unit and the rest going t6 waste.
The orilyq()aglilant tested was fenicchipride (FeCb), thes~mecoagllIant used by the
Marden W~lterTreatmentPlant.Thedosewas varied from 3.,20 ITlglI with the optimal
dose O~Cl1niIig inth~8-12tftgll t3;llge. Acid was also iT1trocluc~ci tQJurth~ritnproveDAP
effl uet1twaterqua.lity.
Throllgoollttbestudy p~ri()dt onIinelIlonitoring of taw and.J:)AF~l1rbidity, pH,ahd raw
di ssolved()xygcn (D())Wer~data logged. lnaddi tion,rolJtin~s~thpIC$Wer~cpnectecl
~ng~t1(J.)y~~qqri"$it~PYJ&qpqlqst~fffor .tQt*I itOJ1,. dis~()~ vC,Q. irPft,tQta.l. .m~gg~9~s~, .atldtrueandaPPilreh tcolobMWIlernplo red an ind~pen dentl3,b6rat9ryfotT()C,1)QC,UV ..
Z54a.Qs.qrpa.t1(;~, Ja.lg3;~ .qQqQt~t ~I1QJil1)i t~cl:q BP te~ ting.;QAffilW1gy s9JiRs. (;()ntetlt (%
. qrysoHds)and D AF$ludg~pr()ducti on '" a 11l111es . were sampled a.ndanalyz~datI..e6po 1 d'Research aociPevelopmentLaboratory.
MontgoPtery WatsOti Haria Company Confid.~t1na1
United Water Idaho
Boise, Idaho
2. Introduction
United Water is cutTentIy evaluating dissolved air flotation (DAF) pretreatment followed
by ultrafiltration membranes as a treatmenLoption for a future plant to treat Boise River
raw water. The existing Marden 81. Water Treatment Plant (WTP) is a direct filtration
plant, whiCh utilizes ferric chloride along With a polymer in its floccu)atioJ1process for
treatment of Boise River water.
The Boise River can experience high algal activity each spring and early slimmer. TheDAF pr6cessis being considered tocotnb~t this algae bloom in the fut~te.For the reSt of
the year, the source water quality is typically of Jaw turbidity 1 ~3NTU, low iron, lowmanganese, and relati vely low color.
F .Leopold Company,Inc.wasipvitedto condl1ct a water treatm~nt stlldyusing its
stale ofthea.rt DAFpilot plantirlthell1onthofAugu~t 2001. The results indicatedjhatthe DAFpretreatrrtentprocess, once optimized, was able to produce enhanced effluent
water quality at high loading rates.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise~ Idaho
3. Objectives
The objectives for the DAFpilot study were set out to help determine potential
pretreatment design criteria for United Water s future membrane plant. These objectives
are outlined below:
DetermineDAF perfoI111ance with respect to chemical requirements, effluent
turbidity, and DAB effluent water quality (total iron, dissolved iron, manganese, true
color, apparent color, TOC, DOC, and UY...254 absorbance).
Confinn and validate the LeopQld DAFpreliminarydesign parameters for a newtreatment facility in Boise, Idaho.
Detennine the design DAB loading rate applicable to the Boise River Source Water.
Characterize the sludge quality and quantity prodllced using the DAFpretreatmentprocess.
Montgomery Watson Harza CompariYConfidential
United Water Idaho
Boise. Idaho
Methodology
The Leopold DAF pilot plant was set up atthe Marden Water Treatment Plant on
Al.1gust 2001. The I.1nit was set up and a 2.5 ingh diameter influenthQsewas connected to
the plant's raw water intake pipe prior to any chemical addition.
TheDAF unit is equipped with two fullyalltomated one-square footfjlter units, Whichwere notutilizedatMWH' s arid United Water s requ~st AJlDAFeffJuentwater was
discharged into a 5Q gallon holding tank, whichproviqed the Jeed wat~rfor the
membrane unit 6perafed by Carollo Engineers.
All instrpmetlts wer~ calibrated onpo\V~r..uP!whichincludedc.iQsiI1gpU1)1pS,
turbidimeters, and pH probes. HachpH 7 and pH 10 buffer soluti onswere used tocalibrate thepIlpr()tJ~saod HachgQNTU forma2:in~f)o)ution was ma.cleup from stockfor calibtationofthe turbidirneters.
On-line data logging included raw wateraridDAF turbidity, pH, and raw water dissolved
oxygen. A Hachj)R 2000 spectroph(jtoJT1~t~r \Vasemploy~dforcol()rtlndI11etal ion
analysis~ Inadc:iition toth.e~et.~st~~.sI4cJge saltlple~\V~tesent to ~()P()lc.i'~R.e~ear~h~11dDeveloprnentLabforpercentagedtysdlids an~lysis.MWH set1tsatl1ple~ tpan
i nd~penc1(mt lab9ra~ory Joe TO.DOC, JJV-254.ab$Qrtn.\nce,Hfrlii~cll):Irp ~t1d~lgq~counts.
M()ntgoni~l"Y Watson Harza Company C()nfidenti~t
United Water Idaho
Boise. Idaho
5. Results
1 Raw Water Quality
During the pilotstudy, the Boise River water sC)urc~contained low levelsof turbidity,
iron, manganese, and color. The source water can develop high levels ofaIgae in the
spring, which significantly effects raw water quality and treatability at the existing
Marden St. WTP. The normal raw water cpnditions,as well as the yearly algae bloom,
suggests thatthis wateris a suitabJecandidatefor PAF.
1 Raw Water Turbidity
Clarity of the Water isirnportant inproducingpbtable water. Turbidity is caused by
suspended andcolloidcil matter such as clay , silt, organic and inorgaIlicmatter andbthermicroscopic organisms, such.asa1gae.
Raw water turbidity was measured usiJtg a Hach 1720D on-line turbidimeter . FigUre!
shoWs therawwatetturbidity from 1O'f\ugustthrough 19 Al.lgust2001. It canheenseen
that the turbidity nmgedfrom 1210 2.5NTU throughoutthe studyaveragingatl.3NTU.
r'
! 5I ~
I ~.
I ~
....
I ~
0 .:. ~
- - ~ - - - .~ - -- -- ~ - - - - - - ~---~- - ----
0000 0000 00.00 00 0000 00000000000 ooQ
L~_
il~i
~~.~ $~..~.
~:.~I~
~!! ~~.~.
..~.m~~.~~~
.....~...~..~ --"'""'.--..". ."""~"".'.,...,..'._-,,-~,"-~"--
R~WWat~rTurbiciity
FigUre 1. Raw.W aterTl1rbidity ~/l OIOltQ8!29101.
Montgomery Watsol1Har.za CQmpany Confidential
United Water Idaho
Bojse, Idaho
Raw Water pH
Raw water pH was measured on a continuous basis using Signet on-line pH meters. The
pH r~adings were constant and predictable on a daily basis throughout the study. The pH
ranged from 7.0 to 8.0 throughout each day due to the photosynthetic effects of algal
growth caused by intense sunlight on the river. During dAylight hOllrs plant
photosynthesis activity removescarb6n dioxide (CO2) from the source Water, making the
waterless acidic, hence a rise in the raw water pH. In theeveuingr~spirationoccurs,
which consumes oxygen in the waterauQ prodQcesCO2, tl1er~by decreasing tl1epH.
Figure2 shows the pH ~rends from 10 August 2001 througl1 the completion of the study.
Figure 3 zooms in on a single day, further demonstrating the diurnalcycJe.
Raw Water
9 .
~ 7;
7 ;1
._-;-, .,--- -.,,- .,,- ...,. . ., .. .' ,.. .' ...- .... -- ----..... -.... - ......,- ..,- '..... - '-- - ..... ---...- ...........-:..... ,..... .....
:R~~gg~~~~~ '~g~g~~~~~ Q.g~~g,~~!i?QQQQ00 .'r-:'('I') """"""Il)Il)(O(O f'o.; f'o.;,CO 'Q)Q) 00 - .C\I ('1'), ('1')""" """LOll) CO(O,,~f'o.; ,co 0)~;:~i::!::: ":0::. ~~t: ,t:, '~t:t:!:::t: .:t,
~~~~ ' ~' ~.'~ ,~.,~~~ ,~ '.~' ~"~ ~
(Ie) OOW W '(I() co COCOCIO coco coco coco co co co 00 00 00 coco 00 coco co co Q)a:)coCOa:)
Date
Figul"e~.J:(awWQ.ter pH SIJO/Ol, to812~/O1!
Montgomery Watson Harza Co mpal1Y Con fidentia 1
United Water Idaho
Boise, Idaho
Raw Water Daily pH Cycle
::c: 7.a. 7.
.-..'...._.-_.._.._.~....... .
7 . .
~. ..-.... '.." ."'..".""
(\I
i\i
C\i
,...
r..:
(")
(ri
,...
.:.tC\I
.:.t
,...
C\j
.q-.....,...,...,...
t\I
Time
Figure 3. Raw Water Daily pH Cycle.
1.3 Raw Water Color
1 Apparent Color
Apparent color is determined using the initial sample without filtration or centrifugation.
Figure 4 shows the raw water apparent color re$ults for the entire study period. The
apparent color varied from 8 to 22 Pt-Co with an average of 16 Pt-Co color units.
2 True Color
The true color of the raw water was measured by filtering the raw water sample through a
0.45 J.Lm OFC filter paper. This aJ1ows a measure of the dissolved color component with
the absence of turbidity or suspended matter.
Figure 5 shows the raw water true color results. These results were significantly lowerthan the apparent color results
(....
70% less), showing that the majority of the color was
particulate in nature.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise. Idaho
Raw Water Apparent Color Residuals
20,
ftI
-".,.-
. ,0 -"~'.~.~"".~'~
.'.~
w.,.~.,~,.,....
..,.....-~....~.----- .. .. ..
e .
.. ...... .. .. ..
e--e:-..~
!,,",,.
.o,"w~
!.~- ...
o .
-_.__."...'._...-."~M..~.'"."...,--
._,-_.,---~~_
w."
,-",..,.~. .
or-
.....................,...................'"'"'
'I'"''I'"''I'"'or-
,...
or-
,...,...,........
t:::
--. ......;:,...........
to-:r..;
.....................................
(\J
Date
.."., -'.'.'.-'-". ._'._--".~------,,_.__..,-_.~_._..- _._'-~-".'."---"'-"_._""-""'~.~--"-,---",-,-,-"'.'-""" ""'."..'.",-,."-,--",,_._-,-,, -'
--'-'w
. .
Figure 4. Raw Water Apparent Color Ilesidu3ls.
Raw. W aterTrueC()lorRe~idu a
'.w...
."._".."",.",..,_"~_.....,,_.._..-----,.~.--.~--....
::s
-.,.....~--.-':--'--_."---._-"..-,,..-,-"""--~~"":~.-,. .'. ...... -.
It .
. .. ......,-:"";-".._",..,.".:;,,,,,.,.:;---,--_.....""',...,........
t:::
,.......'"'"'.....,........
t(j t(j t(j t(j t(j
Date
......"..,.
'1'"'
,...,.......,...
r:.::
.... ....,..,.
r..;
O;j
...-....,.,..,........
:Se
....
Figure RawWa~erTrlleG6IorRe$i(jII~I~h
Montgomery Watson Harza Company. Confidential
...,.-..-.-.......----..--....--,..-.....-----...--""--.. .._-'----------
United Water Idaho
Boise, Idaho
4 Raw Water Iron Concentration
1 Raw Water Total Iron
Iron, in water, may occur in true solution, in colloidal state, in inorganic or organic iron
complexes, or in relati vely coarsesusperided particles. It can be either suspended or
dissolved.
Raw water iron samples collected during the study showed very little total iron. Totaliron ranged between non detect (ND)to 0.1 mgllaveragingQ.O4mg/t These results fallwell below the National Drinking Water Standards maximum contaminant level of 0.
mgll. Figure 6 shows the data collected from the samples taken.
Raw Water TotallronCorlcentration
~. 0.
S. 0.
2" ""',-~,,",..
.!:
:! 0.
~, 0.
~~~'-~~,- -'~'"" .._""-"._'.'.'_.- _._~~ ---, - ~~
MCL (0.3 rrgll)
._.~,_._-,-~, ,.'_."'....,...,...
m.,-...
,..,..,"."."-,.,....,.,......"".,. "'.".."'.'...'.'.."...'",."",,."'.,..,- ".."""'-.,~.~.,.,.-"....,.-.,.."."".."....'."."".".'...._"."-"-_.~-_., ..,.,...".,..,--_.,,-~,..,~---.._,-_..,""...,".",.,
.w,
---_.~-,' .....'..'_.""'.."'.' ."-'_."~-'-~_.~-"--."_.."''--'--'----_._."""'""~ ~ ~ ~ ~ - ~ ~ ~ ~ - ~ -
0 0 0 0 0 0 0 0 Q Q
-..~ '(?j ~
(O a5 (:) C3 .~ CO C) ~
- ~
r::: ~ dS .
~ ~ ~ ~
~ i i g i i ~ S ~ ~ ~ I
~ ~
Date
Figure 6.. Raw Water TotallronConcefitr1jtioJj.
M(jntgomery Watson Harza Company Confidential
United Water Idaho
Boise, Idaho
2 Raw Water Dissolved Iron
Dissolved iron concentrations were obtained by filtering the samples through a 0.45 J.lmOPC filter paper. Any colloidal matter suspended in the grab sample was removed by the
filter papers allowing for measurement of the iron in solutioo. MembranefowHngcan be
significantly affected by the .dissoI ved iron compQnent Particulate ironwHl r~main
, the outside of the membrane fiber and can be easily removed during theb"ckwash
process. Dissolved iron however, can integrate its way into the fiber creating a blinding
effect, which maynQt be easily remQv~d in nonnalbackwash procedur~s. Figure 7shows the raw water dissolved iron data obtaineq throughout the study.
._~""._'.,~".,.,_.,~""'-"_..'."',..,.".,,_.,-~-- ~-'
Raw Water Dissolved Iron Concentration
' 0.
6 0.
.!:::.'.'..'.'."""...'..'...'.'..'-'..-----."-.,.,..",.,...............,-...,.....---.._._--_--_._---_.._.,.-....,...,_....-....... '.
'tJ~ 0,'0 en
is 0,
..".....---_...,....._......;."...._,......,....._._"....-...---_...-_._~._~-.._---.. ~ -~ - - - -- - - ~
0 0 0 0 0 0 0 0 ~
~ ~
Q .
~ ~ ~ ~ ~ ~ -
w ~
~ - - -- - - -
- N N N
- ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ .
Date
0...
.""..""..".....~....
Figure 7 Raw Water Dissolved IronConce~tration.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise, Idaho
Raw Water Aluminum
The lack of aluminum in the source water led to reduction in monitoring throughoutthestudy. In 5 out of the 6 of the tests analyzed, aluminum showed up ND.
1.6 Raw Water Manganese
Manganese, in high concentrations, imparts objectionable stains to laundry and plumbing
fixtures. The regulations regarding acceptable limits in potable water stem from these
rather than toxicological considerations.
Orahsamples analyzed on the Hach DR 2000 indicated. that manganese was presel1t in
concentrations below the regulatedH mil of 0.05 mglLFigure8 shows that raw water
manganese'ranged from ND to 0.033 I11g11 with an averag~of 0.020 .mgll.
.--.'-.""'.""""..'.""""'.-"-
~' o.
gO.O4 -
! 0.
g.().
CIS
:& 0.
M:l.(0 O5111V1)
~.-._-_.,.,.,.,.".".",...._....",....,~,-_..,,~-,,--~~,--~-
..... ..... T""
.... .... '
0 b 0
~ ~ ~ ~ .-.------.---.
Water, Mal1g~neseC on c~l1tr~ti()1"I
06.
.~~.'."-C'-
.'.'-~_
;'_~4~
, , """"""".'._';".
:-e._-
'---~~" . . .. .. ..- ---.
tit
. .--"".,.,..,-".,.,,'.,..._.,-....
T"" -
- - ~ ~ - - ~ ..... ~
~ i i ~ ~ 8 8 ~ B i ~ ~ i ~ I
~ ~
~ ~ ~ i i i i j i ci .
~ ~ . .
Date
."._.~..",,--. . .
Figure 8. Raw'WaterMafiganeseConcel1ttation.
Montgomery Watson Hllrza CQ rnp~n yc::onfidentia I
United Water Idaho
Boise, Idaho
1.7 Raw Water Dissolved Oxygen
Raw water dissolved oxygen (DO) ranged between 5.3 mgll to 7.5mg/1. Dropping
slightly through the last 10 days of the study. Each day would show a trend directly
related to the daily pH curve and the photosynthesis process occurring in the Boise River
(See Figure 3). Figure 9 shows dissolved oxygen concentrations throughout the study.
Raw Water Dissolved Oxygen
5 ---..--..
'II r-ffi-l~-r .
.... ...
t.- :
------------; -"". --.---------,----
~ 6.
ci 5.5 --'--.,----
,_.. ------
_m__--..__----.
-...------..--.---
....---- d...
.... -........
T"" T"" T"" T"" T"" T"" T"" T"" T"" T"" T"" T"" T"" T"" ..-
....
.... T"" ..- T"" T"" T"" T""0 0 0 a 0 0 0 e e Q Q ~ ~ ~ ~ e
' g ~
Q ~ ~ g ~ ~
~ ~ ~
in t::' ~ 03 0
..- ~
(D U.;J f."J
'"
"q' W ,~ U.;J ...-co CO en ca as ca as
..- ..-
T"" ......
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~ ~"""
w w
-..
Date
Figure 9. Raw Water Dissolved Oxygen Concentrations.
Raw Water Temperature
As can be seen in Figure 10 raw water temperature varied daily from 62 op to 70 o
following a similar diurnal cycle to thatof raw water pH and DO.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise, Idaho
Raw Water Temperature
69 ---
II.; 67-
65 --.c '" --- -Q) 63" -
-.,-_., "-"'--
61".-
-- ,
59 """'-""""""-
'-'
57
" ----------"'" ,-",- ", """"',.,-, '------"'-'--"---"'---------"""--""""--"---------------'---------'"-----------------'----------------------- - - - - - - - - - - - - - -- - - - - - -- - - - -- - - - - -
!2 e e e ~
~.
e-.~ ~ Q Q ~Q e ~ ~'
g ~ ~ ~.
R ,~ Q00- - ~ ~ ~ ~ ~ ~ w ~ ~ w ~ m a a - - ~ ~ ~ ~ v ~ ~ w w ~ ~ w ~
~. ~ a5 ~ ?a .
~ .~ ~ ~ .~ .~, .
as' .a;
~ .~ ~ ~ .~ .~ .~ '~ ..~ .~ .~ ~ ~ ,~ .~ .~ "~ "
Date
Figure 1 O. Raw Water Temperature8/! % 1 to 8/29/01.
Dissolved Air Flotation
'I'he~()pold DAF pilot unit was in.itially set up with .3 DAF'. loading rate of4gpI11lft
with 20 minutes of contact time in the flocculation tanks. The loading rate was later
increa$ed to8gpJ;nlft redllcing th~ fJoc;culation time to lOfuihutes forthese~oJtd half()f
the study. AIIDAF effluent was touted directly into a holding tank which fed the
ultrafiltration membral1epilot unit.
Theretycle rate was un stahle for the first five days of the study. Problems with the PID
control algotithmwere corrected~ndlb~r~~ycl~r;lte Wa~ set ~t 14 percent pf t6talfIowfor thefirsthalfofthe study. Jnthe re1T1 Ciifti fig portion of the study,theDAF tecyc1erate
was setat6% with 'Ct.two day test at 10%.. Typical full"'-scale r~cycle d~ign flpw ratesare between 7 and 12%.
Ferric chloride wasthe only coagulant used throughout the stlJdy. Doses were varied
from 3mgll to 20rIigllin order to determine optimal conditions. ThemajorityofDAF
testing was coriductedin the 8-12 rhglJ range.
fIydrQchloric (HCI) at)d.sylfuric(H2S04) acids were incorporatedto l()\Verthy'P~ of the
raw water. Theflocculaticrnand coagulatibnprbqesswithferricchlorideis oft~fltnote
effective when the pH of the system is lowered. Acid Was gradually addedf ata variety
of dosing rates,ioordeftod.eten11ine theoptit11alrange fdfeffectivecoagulatioh with
Montgomery Watson Harza Cornpany Confidential
United Water Idaho
Boise, Idaho
ferric. The optimum pH in terms of producing low turbidityDAF effluent water was in
the 6.0 to 6.1 range for this study.
DAFTurbidity
The turbidity data from dates 10 August 01 to 29 August 01 is shown in Figure 11. Thedata presented below includes changes in loading rates, coagulant dose, acid dose, and
recycle rates. Figure 11 shows several peaks in turbidity, as system breakdown points
were established and less than optimum conditions existed, defining th~. process envelopefor theDAFpenormance under those conditions. In general theDAFeffluent turbidity
was held between 0.5 and 0.7 NTU fot the majority of the study.
---""-""""""'~'"""'-"""--'
RaW and DAF TLirbidity
::)
~ 1.
" ...,-.".,",-'"""...-.
l:'
' 1
... 05
:".;..,...
'I"" ,'I"" '1"", 'I"" ,...
,... ,
r- ...... :,r-
",
"":,r- ,
"""""""""
T'"' ""
""'" ,
'r- ,... """"
, ,""'" ,... :
"" '1""'
"" '
'I""e~Q.~
~ .
Qge"
.~,
S2e_S2"
~g . '~ ,
00
-.- ,...
c;'J 'oCf""fIO,LO ww I'I'W ,en 0 0....
....
1;',/ \')~'J "q" "q"1t) ,w ,q,J1' "r--; ,g;)C)a; ~ 'fij ,
" '
a; ,.~a;
~ ,
a; ,?i; ~ ,
~ ,~ '
a:i ,
$~ :~' ~ "~ '~- ~ :.~ ,~ '
ta '
~ '" ,, :~ ,
Date
I ~DAF
===-
R;;;-'
Figure 11. Raw, andDAF Turbi~ity8/10to8t29.
2 DAF
pH proved to be an important factor in the pilot study. Raw water pH varied from
appr6ximately 7.0. to 8.0. thtoughollt the day withfheDAFeffluehtpH varying accordingfetTicdose. Ferrig chlorid~isacidic in nAture, which atthe optimal dosingtflte(12
mg/1) dropped the pH front, 7Ato6.7. Fora ll1~xim-Qrii fyrricchloridedose 20mg/l,
th~plIdr()pp~d from~QQ1it7.qtQQ.5 ,on .average. ,()ri9 AugllstOlaeidwas irit(od11ged
fo, the system aheadofcoClgl.llantaddition. ThepHWasil1itiallyadjusted usingfuurlatic
~pi(r(31.4~%IIGI)' ~hd .tat~ri $it:lg~93""Q(5%~olll~iQn9f~tllft:41c aCi4~, ,Jfwasdisc9:v~r~difithepilotstLidy that an optirli*111pfItange for tbe~oflg1.ilai1tbeiriguti1itedwas 6.
()j~.
Adding extra acid todrivejhe :pAF'~rfluent pH below Ci.Odidnptproduce any sllbshltl,tial
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise, Idaho
benefits in DAFtreated water quality. Figure 12 shows the DAF effluent pH from 10
August 01 to 29 August 01.
DAF Effluent pH
'."."""""""""""""."'" """"'-"""""""""""""""""""""""""""""""""","""""'.""""""""""""""""""""".""""""""""""""""'.......,_..".
5 - ~~~_..,"_.~....
6 -
....... ,...
"1"'"............
...... ,... """""'
T- '......
...... ......,......,...... ..........'..- .... .........
.......... ....;;-...... ro ............ ro
, ,....~ .
S2",:!;2
~.'
f2"",,-
' ~'~'
QQg,~R 0 ~
.... ......
~ ~ ~ ~ ~ w w ~ ~ w ~ moo.....
.... ~
~ w ~~ W ~ w w ~ ~ w ~
' "~ .,
a; ,a; '
~ ~ ~ '
a; a; ~ ~ ,
~ '~ ,~ ~" ~ ~ ~ ~ ~ ..~ .' ~,~ ,~ .~ '~ ~ ~ ~ '
Date
Figu...e 12.DAF Effluent pH 8/10 to 8/29.
(2Joagulation
Tl1echpice of coag111antcan pl~yani111P()t1antr()~~jntl1ep~ffQi1J.i::\nce.ofany
clariticattol1 process. D AF 'i s achemi call Ybasedclarificati oris ystefuandpfoper
co~g11.1~ntchoic~ is essentiaL Forthfsp#rticlliarstlldy OJ1..1ypneco:iglllilnfwas testedciue
to efflU~t1twatetq liali tyconstrf\i rits'aTltim~IJ1bfanelesting.Thet:o~gt.tlaritchosenW as
ferrlc~tH()rid~( 43%) of specificgt:aYityl.3~~Thisjstl1e&a.mec()~gula.nt thatfheexistingM~d~n .~t WTP cunefltly uti1i~~s. "Figure', shows.'lh~~Q~l~lllant(I()$e pptiJ11ji~tjQtJ
curve at4 gpmlft2 . Th ec 0 rv esl1 0 \V s1 hat dirnhil shi f1 gte turns tifiterrn sofDAF etflQ~fitttitbidity)wetetealized forfenic chlbridedose$ above 10 to 12mglL
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise, Idaho
r-----'------'-"
"-' ---"'-''-------
DAF Effluent Turbidity YS. Ferric Dose
Loading Rate = 4 gpm/ft2
S' '
.. --"-,-"",."....,._-,,,---,,"._--,
!z
.......
:E=' '5 --~.._------"._-".,,"_...,
.,_.,,---
"0
:is
...
?! 1.5 -.'--""""-,--,."
"._,--",~-,,,
5.. ,
.,-"'---""._-----""--"""-'" """""'" "--"--"""--""""'.'~"'--'-""'-""'-',--,--- .,- ..,"'--------.---..,-..---...--,..--"-----."-------..,,..------------,--'."""'--"""'""--"",.."...,, ,,--,-"""""-""""""",,""""",----"",------..'_..,--"-,,,-"~"---'-"""'-'-"--"""-'-"-'--'-'--"'-.""-"'-'-'---,-,,"""'-'" ----""""""""""-.-""---
FerricChlorid, Do$~(mgll)
F'igore 13.F erric Optintiz~ ti()nat 4gpRJIft with out pH Adj ustItlent.
pH was soon aftetdiscovered 10 have a significant impact on .,the effectiven~ss 6ftheferficchI6tide. Figure 14 displays~h~pH ()pthllizationclu'Ve whichdifectly illtpacted
th~ remainder of the stuclY .Thela~tweek;ofth~studyattl1efirst Ipadingrate pf4
gpfulft an d tb e~oti resecoo dhal f()f'th~istQdYw~&funwith the pH 1n1 t'S6pti ma.lra.nge
of 6.0 10,. Figure 15 displays a fetricd6seoptiT11izationcurve at 4 gprhlft under ideal
pH cbnditions.Figure 16 shQwsa femcoptimiz(itioJ1cutve with the ~llrface lo(iWflgtateincreased t()8gpm)fi2
. ,
:Montgomery Watson Haria CorTIpany. Confidential
United Water Idaho
Boise, Idaho
-~-'--""""""-"'" .'" ." ". ... """""""""""""""."".""""".."......"'.,."""","""""-"""""""""""""""""""""'""--""-"~""
DAFEffluent Turbidity VI.
Dose: 12 mg/l Ferric Chloride
7 '-"""""""
-""""'"' "
""""'W_"""'
--"'-"'~'-~-"'-'_.....",-,.....",....
3 '
"---"-"""'-' """""""""""""..,,"""" "" """""""""""."'" "'~'.-"" """""".._-..-.-....~------
a .
_~~""""-"---
8 6.
--,__.
0..6.4 6.
DAF Effluent pH
Figure 14. Turbidity Optimization with pH Adjustmetand FeCI3 Doseof.12 mg/l.
5."
" 0.4
;g.. 0.3'_"."M"_""'-"'~-
j ~ 0.2.
'-"""-"'-'-
OAF EffluehtTurbidityvs. Ferric ()ose
pH htOpthnal Range (6.0~6.1), SLR= 4gpmm2
"-'--"""'....-.
..-"....".."..""",........"m"",..""..-......
..,,-"""-..---.---
1 .. """
"""""'--""....-.-....-,............."..,...-..,.......".....-.....,,_...,.--
Ferric CtlloricJe Pose (mgll)
--_""~~----_..,,,.
Figure 15. Turbidity OptimiiationwithFeCh DoseAdjustmentandpllAdjustm~nt.
Montgomery Watson Harta Company Confidential
United Water Idaho
Boise, Idaho
'-~-'-'--'-~-'--'-'"-----'-"'-"-"--""
-4'_'","""""'.'--
_"""""'-". '.. .""'"""'."."""""."""'--""""""'------"--""
DAF Effluent Turbidity vsFerric Dose
pH(Ei.1); SLR=8gpm1ft2
7 -- ----
--"".""--".,......._..,,.-----
j - 0.6 " --.~j 0
:a 0.
...
::s
to- 0.
,-----" '-""""""""'--'
10
~rric qhloride.lJose(mgl1)
--~--- ----,.".....",.,..~"",.._,...",..-.".,_..,"".....".....,'-""~"""'---""-""""
""""""'-""-", I
13
~-'-"--"'--"""""..--_.._.__.._-,....,'.."-,..._,,..,._~----------.""""""-"---""--""-' ""., ..,""""-"'~._-
Figure!6. Turbidity Optin1iiatiQu with FeCh Dose AdjustmentaildpHA~jtJ,stm.gnt.
. . ', .
5..2~4 Floccu.latlonEl1etgy (0)
TJieLe9poldDAFpiloltraileri~eqtiippedwith2 hOti~()fltitJ, pi(;k'~d ,fence style,
floqqJil atio np addle s . AdJu&tingt h espeedofthesepaddles (ipm )canbe.anim PQrlantf actqt .. in
. .
tt\~ .pptilnizati ()f1pr()~~ss;plp~~1.U~qQ~en~rgYCI.WY~sW~fcg~vel()p~4f9I'. th~
DAF' .systel11 under stable conditions athotl1 .iQadifigr~t~~(4gpfi1lfi2 artd 8gpm/rrJ.. The
rangydbetw~en 50 OOOand IQQ~QOq.MCia~Rgp.1'tilft from4.Q,Q(lQto47,OO()..
f1occulatioTl~nergyatthe .lowerenrloftn~:t3nge& Was selected TQrenergyconsetvati on
ah(icbst effectiveness.
Montgomery Watsonllitia Company Confidential
United Water Idaho
Boise, Idaho
1 ..
S- 0.
! -
i ~
:a 0.4,
~ 0.
. .-.----""--""-"""""',-.-..--,-'--'--.'
DAF Effluent Turbidity vs. G*
Dose = 12 mg/l Ferric; SLR = 4gpmltr
--".-...,..-........,.,.,.."., '..... "-""-""'~-""""'."-' "'" """"""""""""'" ""- """""""'.""'""""""""""-,.,,,,---.---""',..,....,......,..",...--""""""""""'."""'--- --- -"-".-" - -, - ". "'-"- '-----.." ~~-, --.. ". .,.. ,...."..",..",......,-~.--'.'"."."""""'--"""-""'-"""""'-"-."'""-""'-""".',,--,---'~_'_-,.~,...."~,_.,."..-....,..,..,,._,...., """"-""""-""'---
20000 40000 60000 80000
G*t
100000 120000 140000
.._m
,_.....,.-..--....--"...!---"""."""".""""""""""'
Figure 17. Flocculation Energy Curve; (SLR= 4 gptnfft2
...... 0.
::)
!i' 0.
3i!
.a 0.
'-"-"'-".."'.'.'-'---...
?! 0.
----.-..,......-.,.,..-,......,.....-..,.--
20000
---' .".".,........-...
DAFEffluent Turbidity Ys.
pH(6.1);sLR= ,8 gpmlft2
-"'_"..,.-"..,..' ,-...,,.."".""",.........",....""..,.,,~,"-,--""""..-,---,.., '... ""-""".""-,-"",,,-".."",....."",...
3QO
~_._",,
~ ~t
45OQO
-,,-,
50900 .
. ,
.. 95000
Figure l~. FI()~c1l1Iati()nEnergyCurv~;'(S~Jl=f$ gptNre).
25000
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise. Idaho
Color Removal
1 Apparent Color Removal
Apparent color in the DAF effluent, with the addition of feme chloride, closely followed
the raw water results. The high DAFapparent color can be explained by the large tota)iron residual earned Qyer into the effluent. Figure 19 shows the raw and DAF apparent
color results.
Raw and DAF Apparent Color Residuals
,.!.
a. 30.
-,.-
25-
8 .
5 .
'._---......_._.-._~.,_._..__.,..,.,,-,._-,.,_..._,.,-,"'.'."'-"""'..- ".".-'."."""""-"'-"""--"'---'-'-~------.... -.,;...;,...
~,c........JI!
. ','_..._..,',..,-_....,~..."'"'--"-"--"....'.-",...-..--- """" """ """"--. """"""""",,"',,,_..' """"'-"'" ""'"
M._.,-.--.----..-.,
----,,-.---.....
T'"
..............
T'"
..........
T'"T'"T'"T'"
....-,....,.....-,....,........
;;;r f'.:T'"
......................
T'"
........
iX5 iX5 CX5 Cl5 CX5
Date
'RaW_DAF!
Figure 19. Raw and DAF Apparent COlor.
2 True Color Removal
True color is the dissolved fraetionofthe colorinthewater; DAF effluent troecolorwas
consistently 'lower than rawwat~rtruecolor, which was well belQw water quality
standards (See Figure 20). The average fawwater true ,color was 5 .OPt-Co and the
National Drinking Water Secondary Standard (NDWSS) true color list issetatJ5 PI-Co.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise, Idaho
Raw and DAF True ColorResid~als
8' 10
I- 2
----,-_.,~-.-,'- ""--"""""""-""~'-'--"-"-"'-"---"-,----,,-,-,---, ---,.",-"-,.,,,,,,.-..-,.-,_... .. .....,_..,_._-",.."",".,.....',..",, '..... . .. . . .. .
-8---------.", "-
."--'
..--8---e--_"'-
~...
"'lf-"
---,,.....- '-".'--~--
8-11~'
... . . . . ..
0 '
..- ,...~ ~
co
~ ~ ~ g
to ~
......-,...,...,........,...,....
T'"
.....,...
r...;(Xj
,...,...
Date
~~-~..~~~
Figure20. R.3wandDAF True C()lor.
6 DAF Iron Concentration
Ir9n~oncentrations increa~ed through the DAF proce~s~ In orqertor~mqyettltb~dity and
other constituents, femcchloridewaS Jltroduc ed into thesysteni. Iron ftofilthis
pheroicalqarried oyer in' bothparlic111ate. form and to a lessereJ(;tel1iin diss()I ve4fo1111"
Figures 21 and 22 sho\VtheTesWtsifr()m~hecntite study. Fenic9hI()ride.clos~$Wefe
changed throughout the study ,whicl1effectedthe concentrations of iron through the
system. Thel11aximtlmcohtatninahtl~VelfortQtaliron infinjsl1f~clwa!~fis O.3mgll.
This could easily be satisfied with furthetfiltrati6n.
Montgomery Watson Jlarz Company Confidential
United Water Idaho
Boise. Idaho
------------- - --------------
Raw and DAF Effluent Total Iron Residuals
8 .
1.6
1.4--
.s 1.2 ,-----_
. .
- 6 1 --,--,----
...
= 0.
~ 0.
~ ---,,--,,------ - -,----,-----------"----,---"."..--,--.'---.------.."..--..-",---.--.-,,------.-,,---------..--,,--- --,----,-" --"- , - - - -" _
_.m_
.-___----.. ... .__..,-....,,-'_.._'.-'------. .. .. .... .----------""".., ""-,,-,-,,----"-"----,---~--..-.......-.....
i;5
......................-.....
tXi
Date
-'---_.--"-- p---......................-......................-
C\I C\I
L..-I- Raw.
OAF I
.'I'igure 21. RawandDAF Total Iron.
-.-----.-"'-."-""---- ,,-~--------
Raw and DAF Effluen~ Dissol"edlron Residuals
0.12
E 0.
------,,--_.--~"
c~-,II-~~~.
. .-'-----. - --- --"-~,, -'--"- ..'----- ';";'----""'-.""........ ..--".__._----,~ .""""-,._~~
0 0.
i 0.06
;:to
Ut 0.
is 0.
--- .....
co
.......-.........................................................-..-..-..........(:)(:)(:)(:)..................... ........... .......... ..... ..... ..... ..... ...... ...........(:)-.::'.......-.................
O:J
..............,..............,-.....
('oJ C'.I ('oJ C'.I (\J /;"oJc:o
..........................""""""...........------...........---
CX)
Pille
...- ,
e---_,
.:----------:,:",.--,.-
..... .8
--'-:--~.-";."'--
, 8. ,
--.------.",--,--.,.--,---,-,---;;::'-
fIA;r'
--'--.-""--"-,,,, "'" -- '_--m'
:"'" -----'----""--"--"--'---:-"-'---"---"-'""..""""------"",-""-",,,,,-,,--,,-,-"--------'-----,--------------
Figure 22. Raw and I)~FDiSsolvedlron.
Montgomery Watson Harza Company.C9nfidential
United Water Idaho
Boise. Idaho
7 DAF Manganese Concentrations
Manganese (Mn) concentrations across the OAF were slightly higher than those from the
raw water. Raw water Mn averaged 0.02 mgll with DAF effluent Mn results averaging
03 mg/J. This result can also be attributed to ferric chloride addition. Ferric chloride
contains trace amounts pf manganese. Even with thesJightincrease, Mn levels were still
below the maximum contaminant level of 0.05 rngll(Figure 23).
--~-""""-"",.,,'-,..---,_.-'----..,""--,-.,, "',.---,, "'-"""""--""""""'., '
" - ""."W-
---~
Raw and OAF Manganese Residuals
1 . -----"----
'-"---""'" "..--"---,
E 0.08 --
-'--'--".-",..."""",._-_.....---,.."""-,,,,--
= 0.
~ 0.04
- .::.-.--:---_.""""---"",--""""""":""."",."",,,.""".
"",'--,_cc""""",--
""""" '."""'-'."".'.,,---------'
ca
. -' " . ' - .-.-
:i
, - " ..' '- .
8 -
" '
1.-
-'
0 02
. .
w-"'a-a,,-----
. -.----
e . .
. -----,,"-""""--"""--"-'--, '---""""'--"..-'" '.""'.."-
= O 05rrg/l
.. ...... .,... ..... ....
0 0 0 oi'::' ~ co as
~ ,~ ~ ~ ~ ~ ~ ~ g ~ ~ ~, ~' ~ ~ ~ g ~ ~~ ~ ~ ~ ~ ~ ~......~ ~
N ~ N ~W ~ W ~ W ~ W ~
~ ~
Date
.(RaW.OAF
Figure23. ~aw and DAFMa"gal1e$C.
Montgomery Watson Harza Company Confidential
United Water Idaho
Boise. Idaho
5..8 DAF Sludge
The sludge created by the DAF system floats to the surface of the tank where it builds up
into a sludge blanket, often called "float . This is kept on the surface until it is deemed
necessary to remove it. The pilot plant uses a chain and flight type mechanical skimmer
to pull the sludge over a curved beach for removal. This isJhen captured in a container
where the volume generated and dry solids can be sampled and recorded. For the
majority of the study the skimmer was set-up to run every 90 minutes for 5 minutes at a
speed of4 ft/min.
2..1 Solids Content
Samples of sludge were collected throughout the study to determine the total solids
produced by the DAF process and its settling characteristics. Due to handling and
disposal cost, sludge characteristics are often asimpQt1antasthewaterqmility aspects
the prp!'osedtreatmentplant. A high soli(is concentta.fiQn win reduc~ the costs
disposal significantly. Table 1 shows the results 6fthe percentage dry solids (%ds) tests
that were cam~d out by Lcopold'& Research andPevelopment Laboratory staff.
Minifitu.tri
Maxim 11111
AV'era
5..2 .Sludge Production Volume
During the course of the study, sludge volurnetests werecarriedc)utinorder to detennine
sludgeprodllctionby the DAF process. !heamoJ.lntof~lucige generated will clirectlyimpact the costofciisposaL The volumes wereavera.ged frolD severaldesludge occasions
so accurate prodllction volumes could be determined. Table 2 shows the results of these
tests.
T~bl~ .2.DAF 511.14 .el!r9dlJctip..VpJqm~~(%,:QttQt3IFJ():w).
MJiilriiuni
Maximum
Avera
Montgomery Watson Harza Compan y Confidential
United Water Idaho
Boise. Idaho
6. Conclusion
The pilot plant study using the DAF process proved to be an effective and reliable
operating system for treating the Boise River raw water source. The Leopold DAF unit
met or exceededal1 intended water quality goals.
Ferric chloride was successful in removing the turbidity and true color as well as
producing effluent water with a low dissolvedironcoIlcentration. For the Qnemonthduration of this study the optimal range of fetl"icchloride was found to be between Sand
12 mg/I. An adjusted pH in the range of6.0 to 6.1 was found to be optimal Jorthe
chosen coagulant. Muriatic acid (3L45%HCI)ai1dsLilf1jncacid (93-960/0 H2SO4) were
both used independently to lower pH before injection. of the coagulant. . Turbidity
averaged 0.55 NTU throughout the study teaching a minimum of 0.21 NTU at higher
coagulant doses.
TheDAF perfonnedwell at loading rat~~ of both 4 and 8 gpm/ft:Z A fenicchloridedose
of8mgllat 4 gpm/ft and 10 mg/I at 8 gprnlff was needed to hold the turbidity in the
range 0.5100.6 NTU.
The recyclesystel11 can be designed as a stal1dard packed tower satpratipn!)ystem,.
utilizing efficient dissolution of the air into the water anc:lminimiz;ingpllrnping costs for
the system. The recommendedrecyc1e rate would be 10 % with a bufferedmarginof8 to
12 % design.
Sludge solids averaged 1.81 % dry solids with sludgeprodllcHonvolumesaveta,gil1g
04% of plant flow. The highinfIuentwatetqualityreSultedinl6wersolidsprodtiction.
In general, the system perforfuedvery wen~ ThesYst~l1ite~1edjs c~Pa\)leofprQQucit1gan
efflUent water quality that,w hen followed hyproperfUtration,would. easily. meetcllrrerit
dririking.watetstandards.
Montgomery Watson Harza Company Confidential
APPEND IX B
THE EFFECT OF DAF PRETREATMENT
ON MEMBRANE PERFORMANCE
DAF pretreated water SDS DBP concentrations. However, neither the TTHM or HAA5 levels
exceed MCLs under 48-hour SDS conditions.
Hydraulic Performance of UF Membrane with DAF Pretreatment
(August 2001)
The hydraulic performance of the membrane was measured continuously by on-line
instrumentation on the membrane pilot plant. The parameters measured by the membrane
pilot instrumentation were transmembrane pressure (TMP), specific flux (flux at 200C),
normalized flux, and temperature. The membrane run termination criteria established for the
study was a TMP of 15 psi or a 30-day chemical cleaning interval, whichever came first.
The membrane conditions tested during the first phase of the pilot-study with DAF
pretreatment for the UF membrane were:
Flux: 68 L/m h (40 gpd/fe)
Flow rate: 32.5 gpm
Backwash frequency: 55 min
Percent recovery: 93%
The membrane hydraulic performance data for the first phase of the study is presented
Figure 1.
During the first eight days of operation , only a slight rise in TMP was observed. However, the
TMP did vary significantly (between 5-psi) during the production cycle (between
backwashing). Following backwashing, the TMP remained around 5 psi and climbed to about
7 psi at the end of the production cycle.
Due to a power outage, both the DAF and membrane pilot plants were shutdown on day 8 of
the study. From day 9 to day 11 , the operational conditions of the DAF and membrane pilot
plants were not consistent, which is reflected in the data provided for those days. After
operational conditions were stabilized on day 11 , the TMP increased to between 7 and 9 psi
remaining relatively steady to day 16. On day 17 , the SLR of the DAF was increased from 4
gpm/sq ft to 8 gpm/sq ft and a constant increase in TMP was observed from day 17 to day 25
of membrane operation. The approximate fouling rate during this period was 0.psi/day,
compared to the relatively stable TMPs observed at the DAF SLR of 4 gpm/sq ft.
Under the conditions tested, the TMP reached only 12 psi after 25 days of operation.
Extrapolation of the TMP data to 30 days (the desired chemical cleaning interval) of
membrane operation, the expected TMP would be about 14 psi , which is less than the
maximum cut-off criteria of 15 psi. It is noteworthy that the maximum recommended operating
pressure for the membrane is 29 psi. Terminal TMP of 15 psi was selected for consistency
with previous testing. However, it would have been possible to operate the membrane up to a
TMP of 29 psi without compromising membrane integrity.
December, 2001
OA
F
Ba
l
.
e

R
i
v
e
r
La
g
o
o
n
Da
t
e
Da
y
Ac
t
i
v
i
t
y

De
s
c
r
i
p
t
i
o
n
Fo
e
d
t
a
D
A
F
Fa
a
d
Ra
t
e
On
s
l
l
e
(g
p
m
)
(g
p
m
)
8/
1
1
0
1
Se
t
-
u
D

IP
a
w
a
r

an
d

W
a
l
a
r

tD

DA
F
8/
2
1
0
1
Se
l
-
u
D

&
D
A
F

Pil
a
t

Te
a
t
8/
3
1
0
1
Se
l
-
u
D

&
D
A
F

Pll
a
l

Te
s
l
8/4
/
0
1
Sta
r
t

1
4
-
Da
v

Te
s
l

1

:
3
0

PM
'
8/5
1
0
1
En
d

o
r

In
i
t
i
a
l

TM
P

Ra
D
l
d

Ri
s
a

fr
o
m

8/6
1
0
1
Bm
0
1
DA
F
O
f
f
f
r
o
m
0
9
:
0
0
I
D
1
2
:
o
o
(
B
)
B/8
I
0
1
DA
F

ra
n

I
r
a
m

4P
M

-
9:
3
0

AM

al
1
2

n
,
B/9
/
0
1
Lo
w
e
r

H
I
D

-
6
.
4-6
.
5 w
/

Mu
r
i
a
U
c

ac
i
d
8/1
0
/
0
1
Se
'
e
c
t

lo
w
e
s
t

no
s
s
l
b
l
e

le
r
r
i
c

da
s
a
n
e
.
8/1
1
/
0
1
81
1
2
1
0
1
Po
w
e
r

Ou
t
a
a
e

al
2
2
:
0
0
.

DA
F

& M
e
m
B/1
3
1
O
1
DA
F

Sta
b
a
l
i
z
e
d

f6
)

12
D
D
m

Fe
r
r
i
c

bv

1
8/1
4
/
0
1
Ac
i
d

Lin
e

Bl
a
w
o
u
l

al

1
9
:
0
0

-
lo
s
s

o
r
B/1
5
/
0
1
Ac
i
d

Lin
e

Fix
e
d
'
"

-
1
0
:
0
0
'

Ra
w

We
l
8/1
8
1
0
1
6/1
7
/
0
1
Sw
i
l
c
h
D
v
e
r

fr
o
m

HC
!

to

H
2
S
O
4

f6
)

B/
1
8
/
0
1
En
d

1
4
-
D
a
y

Te
s
l

(
g
j

13
:
3
0
.

Ac
i
d

Le
a
Sa
m
D
l
e
d

al
1
3
:
4
5

Tu
r
b
l
d
i
t
v

-
0
.
4
N
T
U
s
B/
1
9
1
O
1
B/
2
0
/
0
1
Sta
r
t

or
9
-
Da
y

le
s
l

(
g
j

OB
:
3
0

(g
j

9p
m
8/
2
1
/
0
1
No

D
l
s
l
i
n
c
U
v
e

TM
P

ri
s
e

a
l

be
g
i
n
n
i
n
g
67
.
8/
2
2
/
0
1
67
.
B/2
3
/
0
1
1 h
r
.

sh
u
l
d
D
w
n

al
1
0
:
3
0

AM

ra
r

n
e
w
8/
2
4
1
0
1
Al
k
i
n
=
3
4

pp
m
:

A
l
k
o
u
t
=
1
5
.
8 p
p
m
B/2
5
/
0
1
Re
d
u
c
e
d

FID
C

Sp
e
e
d

to

1
1

&
1
0

rp
m
B/2
6
/
0
1
B/2
7
1
O
1
8/2
8
1
0
1
B/2
9
1
O
1
En
d

o
r

9
-
Da
y

Te
s
l

1
1
!
1
4

PM

Dn
Bo
i
s
e
81
3
0
1
0
1
Pa
c
k

U
D

DA
F

Tr
a
i
l
e
r

fD
r

Sh
i
D
m
e
n
l
B/
3
1
/
0
1
DA
F

Tra
i
l
e
r

Sh
i
n
n
e
d

10
Ne
x
t

Pro

o
c
t
9/
4
1
0
1
9/
5
/
0
1
9/
6
1
0
1
9n
l
O
1
9/
1
0
/
0
1
Ta
b
l
e

A
-
Pri
n
t

D
a
t
e

= F
e
b
r
u
a
r
y

13
,

2
9
9
2
Un
i
t
e
d

W
a
t
e
r

I
d
a
h
o
Co
l
u
m
b
i
a

W
a
t
e
r

T
r
e
a
t
m
e
n
t

P
l
a
n
t

P
i
l
o
t

S
t
u
d
y
DA
F

&

U
F

O
p
e
r
a
t
i
o
n
a
l

P
a
r
a
m
e
t
e
r
s

a
n
d

A
n
a
l
y
t
i
c
a
l

R
e
s
u
l
t
s
TB
D

= T
D

Be

D
e
l
e
n
n
i
n
e
d
.
A
l
l

D
B
P
'
s a
r
e

S
D
S

4B

ho
u
r

h
o
l
d

tim
e

sa
m
p
l
e
s
.

A
p
p
l
i
e
d

Ch
l
o
r
i
n
e

Do
s
e

= 2
.
0 p
p
m

w/
s
t
o
c
k

ch
l
o
r
i
n
e

so
l
u
l
i
o
n
..
V
a
l
u
e
s

E
s
t
a
b
l
i
s
h
e
d

by

C
a
r
o
i
l
D

En
g
i
n
e
e
r
s

(7
/
3
1
/
0
1
)
.

(1
.
lh
r

=

0
.
59

g
p
d
/
f
t
SL
R
Fe
r
r
i
c
Ac
i
d

F
e
e
d
OA
F
OA
F
on

O
A
F
Ch
l
o
r
i
d
e
Ra
t
e
Ac
i
d
EF
F
L
U
E
N
T
Flo
c
c
u
l
a
t
i
o
n
Re
c
y
c
l
e
(g
p
m
l
f
t
'
)
Do
s
e
(m
l
/
m
l
n
)
Da
.
e
E..
.
'
9
Y

Fla
w

R
e
t
e
(m
g
l
l
)
(m
g
l
l
)
('Y
o

01

F
l
a
w
)

BO
00
0

1
2
Un
s
t
a
b
l
e

3
60
0
0
0
1
2
Un
s
t
e
b
l
e

3
60
0
0
0

(
2
)
Un
s
t
e
b
l
e
l
3
12
1
D

00
0

(
2
)
12
.
51
7
1
15
1
D
2
0
l
D
1
2
60
.
00
0
1
2
12
.
12
1
0
1
6
+
A
c
l
d
60
0
0
0
1
2
12
.
60
0
0
0

2
12
.
60
.
00
0

2
12
.
00
0

2
12
.
60
0
0
0

2
12
.
8+
H
C
I
60
0
0
0
2
12
.
B+
H
C
I
60
0
0
0

(
2
12
.
+H
C
I
60
.
00
0

(
2
12
.
B+
H
C
I
60
.
00
0

(
2
)
12
.
+ H
,
SO
,
19
.
60
.
00
0

(
2
)
12
.
B'H
,
SO
,
19
.
00
0

(
2
)
12
.
+ H
,
SO
,
18
.
4
00
0

(B
)
+ H
,
SO
.
18
.
4
00
0

(B
)
11
.
Bt0
1
0

+ H
,
SO
,
17
.
30
.
00
0

(B
)
11
.
0/6
.
OID
1
2

+
H
,
00
0

(B
)
12

+

H
,
SO
,
00
0

(
9
)
12
'

H
,
SO
,
00
0

(
1
0
)
12
'

H
,
SO
,
40
,
00
0

(
1
0
)
10

+

H
,
SO
,
00
0

(
1
0
)
10

+

H
,
SO
,
40
.
00
0

(
1
0
)
10

+

H
,
SO
,
00
0

(
1
0
)
Me
m
D
r
a
n
e
Vis
c
o
s
i
t
y
Ptl
a
t
In
l
e
t
Tra
n
s
-
Flu
x

Ra
l
e
Te
m
p
.
Ull
r
a
F
I
l
I
r
.
.
Pr
e
s
s
u
r
e
.

M
e
m
b
r
a
n
e
/h
r
Ca
"
"
c
t
l
a
n

F
l
a
w

R
.
t
e

Me
m
b
r
a
n
.
Pr
e
s
s
u
r
e
la
D
d
l
f
I
'
\
-
Fa
c
t
o
r
(g
p
m
)
-
(p
.
l
g
)
-
(p
s
l
d
)
-
68
1
4
0
31
.
4 I
D
6
68

4
0
31
.
21
0

7
.
B8

31
.
21
0

7
.
B8
14
0
1
31
.
2 I
D

7
.
6B

14
0
1
31
.
2 I
D

7
.
6B

14
0
1
31
.
2 t
D

7
.
68

(
4
0
1
31
.
2 t
D

7
.
6B

31
.
21
0
7
.
6B

31
.
2 I
D

7
.
6B

31
.
Dff

/ N
D

Da
l
a
BB

40
1
31
.
Off

/ N
D

Da
t
a
6B

40
1
31
.
Off

/ N
D

Da
l
a
68

4
0
1
31
.
61
0
9
68

f
4
0
1
31
.
71
D
9
68

(
4
0
)
31
.
1t
D
9
6B

(4
0
)
31
.
71
0

9
.
68

(
4
0
)
31
.
7 t
o

9
.
6B

(4
0
)
31
.
21
0

9
.
68

(
4
0
)
31
.
5 t
D

9
.
68

(
4
0
)
31
.
61
0
9
.
68

(
4
0
)
31
.
51
D

68

(
4
0
)
31
.
68

(
4
0
)
31
.
68

(
4
0
)
31
.
68

(
4
0
)
31
.
68

(
4
0
)
31
.
68
1
4
0
31
.
Ra
n

I
D
r

-
3
h
r
s

o
n

ra
w

wa
l
e
r
68

4
0
31
.
Ra
n

f
D
r

-
5
h
r
s

a
n

ra
w

wa
l
e
r
Ra
w
W
.
t
e
r
OA
F

Pr
o
d
u
c
t

Wa
t
e
r
Pa
l
y
m
e
m

uF

M
e
m
b
r
a
n
.
Pe
r
m
o
e
t
e

Wa
t
e
r
TO
e
DO
C
UV
"
,
Ala
.
e
To
e
DO
C
UV
"
,
Alg
a
e
TO
e
DO
C
UV
"
,
TT
H
M
'
HA
A
S
'
34
0
/
m
l

03
9
/
e
m
28
6
/
m
l
02
0
/
c
m
44
/
m
l
1R
A
W
-
1D
A
F
F
W
-
1P
E
R
M
-
n-
-
19
0
/
m
1
l
4
03
9
/
e
m
(N
D
I

An
a
l
01
8
/
c
m
21
/
m
l
11

p
p
b
37

p
p
b
2R
A
W
-
2D
A
F
F
W
-
2P
E
R
M
-
20
6
/
m
l

13
9
/
m
l
(
4
03
7
/
e
m
(N
D
I

An
a
l
01
7
/
c
m
38
/
m
l
01
5
/
e
m
pp
b
35

p
p
b
3R
A
W
-
3D
A
F
F
W
-
B-
2
8
3P
E
R
M
-
B
-
un
l
i
l

sh
u
l
d
o
w
n

du
e

1
0

hi
g
h

dp

o
n

ml
c
r
o
-
s
l
r
a
i
n
e
r
un
t
i
l

sh
u
t
d
o
w
n

du
a

t
o

hi

h

d
D

Dn
mlc
r
o
-
s
l
r
a
i
n
a
r
Ch
a
m
i
c
a
l

C
l
e
a
n
.
ln
-
Pla
c
e

(C
I
P
)
Fi
b
e
r

In
l
e
g
r
i
t
y

le
s
l

p
a
l
o
r
m
e
d

an
d

l
a
i
l
e
d

-
I
D
c
a
l
a
d

an
d

e
p
o
x
y

se
a
l
e
d

5
d
a
m
a
g
e
d

fib
e
r
s
.
Sw
i
l
c
h
a
d

fr
o
m

ra
w

wa
t
e
r

su
m
p

10

bD
D
s
l
e
r

pu
m
p
s

(
s
a
n
d

wa
s

a

p
r
o
b
l
e
m

fro
m

th
e

r
a
w

wa
l
e
r

su
m
p

clo
g
g
i
n
g

mi
C
f
0
9
b
a
i
n
e
r
)
.

Sti
l
l

ru
n
n
i
n
g

al

Q

..

3
1

g
p
m
50
0

M
i
c
r
o
n

str
a
i
n
e
r

sh
u
l
d
o
w
n

du
e

t
D

w
e
a
d
s
,
'
o
a
f

pe
r
t
i
c
l
e
s
.

elc
.

b
l
i
n
d
i
n
g

sc
r
e
e
n
.

E
n
t
i
r
e

me
m
b
r
a
n
e

un
l
l

wa
s

s
h
u
l
d
D
w
n

lo
r

I
h
e

we
e
k
e
n
d
.
Re
m
o
v
e
d

in
t
e
r
n
a
l
s

DI ml
c
r
o
s
t
r
a
i
n
e
r

an
d

b
e
g
a
n

ru
n

w
i
l
h

ro
w

wa
l
e
r

dir
e
c
t
l
y

to

m
e
m
b
r
a
n
e
s
.

Pe
r
f
D
n
n
e
d

CI
P

Dn
me
m
b
r
a
n
e
s
.

Me
m
b
"
,
n
e

un
l
l

pla
c
e
d

In
l
o

op
e
r
a
t
I
o
n

al

-

3
1

g
p
m

al
-

2

P
M
.
HA
A
S
'
I 9
/
2
5
/
0
1

I
11
2
/
2
6
/
0
1
1
I 1
/
1
5
/
0
2

I
IE
n
d
o
r
9
-
Da
y
T
e
s
l
I
i
!
l
4
P
M
o
n
B
o
l
s
e
~

0

O
~

I
I

0
.
0
I

I
0
.
I
I

5
5
(
3
2
)

0
.
95

2
5
.
11

54

p
p
b
IM
l
d
-
r
u
n
w
l
F
e
e
l
3

ad
d
l
U
o
n

10

Me
m
b
r
a
l

0

I
I

0
.
0
1

2
.
I
0

I
0
.
I
I

6
8
(
4
0
)

0
.
95

3
1
.
20
1
0
2
5

1
0
1
0
2
0

To
e
R.w

W
.
t
e
OC

I

w
"
,
AF

P
r
o
d
u
c
t

Wa
t
e
r

DO
C

UV
"
,

I

Alg
a
e
Po
l
y
m
e
m

uF

M
e
m
b
n
m
o
DO
C

W"
,
.

TT
H
M
'
AI
.
!
!
.
a
To
e
To
e
3
I

1
.
4
I

-
-

Ig
O
2
/
m
l

4R
A
W
-
25

(
b
y

UW
I

Tu
e
s
d
a
y

Sa
m
p
l
i
n
g
)
I
2
.
5
1

1
.
5
1
0
.
O4
0
/
e
r
n
l

-
-
5R
A
W
-
12
-
26

(
C
I

re
s
.

(i!
)

48

h
r
s

= 0
.
51

p
p
m
.
4
I

1
.
4
1
0
.
03
4
/
e
m
l

-
-

6R
A
W
-
01
-
15

(
C
I

re
s
.

I
i
!
I

1
.
5h
r
s

=
0
.
35

p
p
m
03
1
/
e
m
29

p
p
14
P
E
R
M
-
B
1
.
O3
4
/
e
r
n

1
6

p
p
b

14

b
H
=

B;
TT
H
M
=
2
4

p
p
b
. H
A
A
S

=
1
7

pp
b
)

5
P
E
R
M
-
12
-
26

C
I

r
e
s
i
d
u
a
l

al
4
8

h
r
s

=
0
.
20

p
p
m
.

pH

=

7
.
-
1
.
9
2
.
0
0
.
03
4
/
e
m

4
2

b

17

b
H=
8
;

T
T
H
M
=
1
9
p
p
b
,
HM
5
=
1
2
p
p
b
)

6
P
E
R
M
-
01
-
15
(
C
l
r
e
s
l
d
u
a
l
a
I
1
.
5h
r
s
=
0
.
95
p
p
m
,
pH
=
B
IM
l
d
-
r
u
n
w
/
n
D
F
e
C
I
3
a
d
d
I
U
D
n
l
o
M
e
m
b
i

0
L
0
I

0
1
68
(
4
0
)

0
.
95

3
1
.
~
_
20
1
0
2
5

1
0
1
0
2
0
No
t
e
s
:
2-
T
h
i
s

va
l
u
e

wa
s

d
e
l
e
n
n
l
n
e
d

10

be

n
e
a
r

op
t
i
m
a
l

ba
s
e
d

u
p
o
n

ac
t
u
a
l

DA
F

pi
l
o
l

sc
a
l
e

G
I

ru
n
s

a
l

a c
o
n
s
l
a
n
l

do
s
e

o
f

12

m
g
/
I

Fe
C
I
3

(B
/
0
3
l
2
O
O
1
)
.

R,
=

11

r
p
m
s
, R
,
=5
.
5 r
p
m
s
3.

T
h
e

r
e
c
y
c
l
e

no
w

ra
l
e

o
n

th
e

L
e
o
p
o
l
d

DA
F

un
i
t

Is

h
i
g
h
l
y

un
s
l
a
b
l
e

va
r
y
i
n
g

co
n
s
l
a
n
t
l
y

Ir
o
m

2
.
5 t
o

5
.
8 g
p
m

(7
%

10
16
%
)
.

Un
a
c
c
e
p
l
a
b
l
e

Pe
r
l
D
r
m
a
n
c
e

wh
i
c
h

re
q
u
i
r
e
s

Im
m
e
d
l
a
l
e

at
t
e
n
t
i
o
n

by

L
e
o
p
o
l
d
.
4.
B
o
i
s
e

Riv
e
r

wa
l
e
r

qu
a
l
i
t
y

fro
m

Hw
y

21

b
r
i
d
g
e

an
d

M
a
r
d
e
n

Str
e
e
t

in
l
a
k
e

we
r
e

s
a
m
p
l
e
d

we
e
k
l
y

by

U
W
I

fo
r

a
l
g
a
e

sp
e
c
i
a
t
i
o
n

an
d

e
n
u
m
e
r
a
t
i
o
n

as

w
e
l
l

as

T
O
C
, D
O
C
.

(A
n
a
l
y
s
e
s

by

M
W

l
a
b
s
.
6.
D
A
F

wa
s

l
u
m
e
d

of
f

lo
r

3

ho
u
r
s

10

all
o
w

ho
o
k
-
u
p

Df
ne
w

po
w
e
r

su
p
p
l
y

10
I"
,
i
l
e
r
.

R
a
w

wa
l
e
r

su
p
p
l
y

po
l
n
l

10

DA
F

ch
a
n
g
e
d

10

6'
d
l
a
.

bio
w
-
b
a
c
k

lin
e

aro
u
n
d

ra
w

wa
l
e
r

pu
m
p
s
.

Me
m
b
"
,
n
e

Un
i
t

wa
s

o
f
f

lin
e

be
t
w
e
e
n

09
:
0
0

an
d

1
3
:
0
0
.
7.

M
a
t
t

S.
o
r

Le
o
p
o
l
d

go
l
t
h
e

c
o
n
t
r
o
l
l
D
o
p

sta
b
a
l
i
z
e
d

on

I
h
e

re
c
y
c
l
e

",I
e
.

(
s
a
i
d

It
w
a
s

ho
l
d
i
n
g

sl
e
a
d
y

be
t
w
e
e
n

4
.
3 a
n
d

4
.
8
g
p
m
)
.
8.
R
e
v
o
l
u
t
i
o
n

sp
e
e
d

o
n

flo
c
c
u
l
a
l
o
r
s

wa
s

l
e
f
t

at

s
a
m
e

sp
e
e
d

a
s

wh
e
n

ru
n
n
i
n
g

al
4

g
p
m
l
f
t
'

f
R,
=

1
1

rp
m
s
.

R
,
=5
.
5 r
p
m
s
)
9.
R
e
v
o
l
u
t
i
o
n

sp
e
e
d

o
n

fl
o
c
c
u
l
a
l
o
r
s

wa
s

i
n
c
r
e
a
s
e
d

to
R
,

=

1
3

r
p
m
s
, R
,
=1
1

'O
m
s
:

Tu
r
b
i
d
i
t
y

dr
o
p
p
e
d

by

0
.
15

N
T
U
'
10
.

R
e
v
o
l
u
t
i
o
n

sp
e
e
d

o
n

fI
o
c
c
u
l
e
l
o
r
s

wa
s

c
h
a
n
g
e
d

to
R
,

=

1
1

rp
m
s
,

R
,
=1
0

rp
m
s
TA
B
L
E

A.2
AL
G
A
E

S
A
M
P
L
E

D
A
T
A

2
0
0
1
7/2
4
/
0
1
7/3
1
/
0
1
8/7
/
0
1
8/
1
4
/
0
1
8/2
1
/
0
1
8/2
8
/
0
1
9/4
/
0
1
9/
1
8
/
0
1
9/
2
5
/
0
1
10
/
2
1
0
1
10
/
9
/
0
1
10
/
1
6
/
0
1
Ty
p
e
MW
T
P
Hw
y

2
1
MW
T
P
Hw
y

2
1
MW
T
P
Hw
y

2
1
MW
T
P
Hw
y

2
1
MW
T
P
Hw
y
2
1

I
MW
T
P

I

Hw
y
2
1

I
MW
T
P
Hw
y

2
1
MW
T
P
Hw
y

2
1
MW
T
P
Hw
y
2
1
MW
T
P
Hw
y
2
1
MW
T
P

I

Hw
y
2
1

1
MW
T
P
Hw
y

2
1

I
To
t
a
l

A
I

ae

#
l
m
l
31
2
17
5
31
6
27
4
34
0
22
6
19
0
18
9
20
6
14
1
13
9
21
1
41
0
28
0
46
0
96
8
90
2
98
0
12
5
0
15
7
0
13
1
0
79
0
91
0
99
0
-,.
Ac
h
n
a
n
l
h
e
s
dia
t
o
m
Am
p
h
o
r
a
dia
t
o
m
An
a
b
a
e
n
a
blu
e
-
o
r
.

R
l
a
m
e
n
l
An
i
s
o
n
e
m
a
ne

e
l
l
a
l
e
An
k
l
s
t
r
o
d
e
s
m
u
s
re
e
n
Ap
h
a
n
o
l
h
e
c
e
blu
e
'
o
r
e
e
n
Bo
d
o
na
o
e
l
l
a
l
e
Ca
r
t
e
n
a
na
o
e
l
l
a
l
e
Ch
l
a
m
y
d
o
m
o
n
a
s
na
a
e
l
l
a
l
e
Ch
i
a
r
e
l
l
a
re
e
n
Ch
l
o
r
a
a
o
n
l
u
m
na
a
e
l
l
a
l
e
Cl
o
s
l
e
r
i
o
o
s
i
s
re
e
n
Cl
o
s
l
e
r
i
u
m
re
e
n
Cr
v
D
l
o
m
o
n
a
s
na
a
e
l
l
a
l
e
Cy
a
l
h
o
m
o
n
a
s
na
o
e
l
l
a
l
e
Dlc
l
Y
o
s
p
h
e
n
u
m
re
e
n
Dln
o
b
r
v
o
n
na
a
e
l
l
a
l
e
Ela
k
a
l
o
l
h
r
i
x
re
e
n
Eu
a
l
e
n
a
na
a
e
l
l
a
l
e
Gl
e
n
o
d
l
n
l
u
m
na
g
e
l
l
a
l
e
Gl
a
e
a
c
a
D
s
a
blu
e
-
o
r
a
e
n
Ga
l
e
n
k
l
n
i
a
re
e
n
Go
m
p
h
o
n
e
m
a
dia
t
o
m
Go
m
p
h
o
s
p
h
a
e
n
a
blu
e
-
o
r
e
e
n
Gv
m
n
o
d
i
n
l
u
m
na
a
e
l
l
a
l
e
Gv
r
o
s
l

m
a
dle
l
o
m
Ha
n
n
a
e
a
dia
t
o
m
Lv
n
a
b
v
a
blu
e
-
a
r
.

F
l
l
a
m
e
n
l
Ma
l
l
o
m
o
n
a
s
na
a
e
l
l
a
l
e
Me
r
i
s
m
o
D
o
d
l
a
bl
u
e
.
",e
e
n
Ma
n
a
s
na
o
e
l
l
a
l
e
Ne
d
i
u
m
di
a
t
o
m
Oc
h
r
o
m
o
n
e
s
na
g
e
l
l
a
l
e
O/c
o
m
o
n
a
s
na
a
e
l
l
a
l
e
Oo
c
v
s
l
i
s
re
e
n
Os
c
l
i
l
a
t
o
n
a
blu
e
-
o
r
.

F
l
l
a
m
e
n
l
Pe
r
a
n
e
m
a
na
a
e
l
l
a
l
e
Pe
l
a
l
o
m
o
n
a
s
"a
o
e
l
l
a
l
e
Pl
n
n
u
l
a
r
l
a
dia
t
o
m
Pl
a
n
k
t
o
s
p
h
a
e
r
i
a
re
e
n
Po
l
v
t
o
m
a
je
l
l
a
l
e
Po
l
v
t
a
m
e
l
l
a
Pa
l
e
r
i
D
c
h
r
o
m
a
n
a
s
Rh
v
n
c
h
o
m
o
n
e
s
Sc
e
n
e
d
e
s
m
u
s
re
e
n
Sc
h
r
o
e
d
e
r
l
a
re
e
n
Sp
h
a
e
r
a
c
v
s
l
i
s
re
e
n
Sa
l
r
o
a
v
r
a
re
e
n
, f
i
l
a
m
e
n
t
Sa
i
r
u
l
l
n
a
bl
u
e
-
a
r
.

F
i
l
a
m
e
n
t
Sv
n
e
d
r
a
dl
a
l
o
m
Ta
b
e
l
l
a
n
a
di
a
t
o
m
Te
l
r
a
e
d
r
o
n
re
e
n
Te
l
r
a
s
l
r
u
m
re
e
n
Tr
a
c
h
e
l
o
m
a
n
a
s
"a
a
e
l
l
a
l
e
Ula
l
h
n
x
gr
e
e
n
, f
i
l
a
m
e
n
t