Sie sind auf Seite 1von 112

Interim Report

Contents
1.

INTRODUCTION
8
1.1

Background
8

1.2

Field Investigations
9
1.3

Bathymetry
9

1.4

Scope of this Interim Report


9

2.

SITE CONDITIONS
10
2.1

Site Appreciation
10

2.2

Finished Floor Level


12
2.3

Climate
12

2.4

Typical Oceanographic Conditions


13

3.

WATER QUALITY
14
3.1

Seawater Quality
14

3.1.1

Importance of Sea Water Quality


14

3.1.2

Temperature and Salinity Trends


15
3.1.3

Field Investigations

Page 1

Interim Report

15
3.1.4

Water Quality Design Criteria


18

3.2

Water Quality Targets


20

3.2.1

Total Dissolved Solids


21
3.2.2

Boron
22

3.2.3

Jelly Fish Blooms


23

4.

DESIGN LIFE AND AVAILABILITY; PRIMARY FLOWS


24
4.1

Design Life
24

4.2

Operations and Availability


24
4.2.1

RO Rack Configuration
24

4.3

Primary Flows
25

5.

SEAWATER INTAKE
27
5.1

Primary Flows
27

5.2

Intake structure with Screen offshore


27

5.3

Active Screens before the Sea Water Pumps


28
5.4

Sea Water Pumping Station


29

5.5

Shock chlorination system


Page 2

Interim Report

32
5.6

Intake / Outfall Pipes


32
5.6.1

Pipe jacking
33

5.6.2

Main Jacking Station


33
5.6.3

Inter-jacks
34

5.6.4

Slurry System
34

6.

PRETREATMENT SYSTEM
35
6.1

Pretreatment Process
35

6.2

Sea Water Quality / Treatment Processes


35

6.2.1

Key Water Quality Parameters Pre-Treatment


35

6.2.2

The Pre-Treatment Processes- Scheme


37

6.3

pH Correction
37

6.4

Coagulation
37

6.5

Flocculation
38

6.6

Lamella Settlers
39

6.7

Dissolved Air Flotation (DAF)


40
6.8

Ultrafiltration (UF)
43
Page 3

Interim Report

6.9

Ultrafiltration Water Storage


44

7.

REVERSE OSMOSIS PLANT


45
7.1

Overview of RO Plant
45

7.2

Pumping System Design and Pressure Control


45
7.3

Cartridge Filtration
47

7.4

Oxidant Control
47

7.5

Scale Control
47

7.6

Reverse Osmosis Pumping and Energy Recovery System


47
7.7

RO Rack Design
52

7.8

RO Clean In Place System


54
7.9

Permeate Storage
56

POST TREATMENT

8.

57
8.1

Re-mineralization/Stabilization
57
8.1.1

Requirement for Stabilisation


57

8.1.2

Hydrated Lime & Carbon Dioxide


58
8.1.3

Key design Criteria


59

Page 4

Interim Report

8.2

Disinfection
61

9.

CHEMICAL SYSTEMS
67
9.1

Antiscalant
68

9.2

Carbon Dioxide
69

9.3

Chlorine Gas
69

9.4

Ferric Chloride
71

9.5

Hydrated Lime
71

9.6

Flotation Aid Polyelectrolyte


72
9.7

Sodium Hydroxide
73

9.8

Sodium (Meta)bisulphite
74
9.9

Sulphuric Acid
75

9.10 Ultra Filtration Chemicals


76
9.11 Miscellaneous Chemicals
76
9.11.1

Cleaning-in-place (CIP)
77
9.11.2

Flushing
78

10. WATER STORAGE AND TRANSFER


79
Page 5

Interim Report

10.1 Process water storage


79
10.2 Process water transfer pump station
79

11.

SEAWATER OUTFALL
80
11.1 Outfall Pipeline
80
11.2 Diffusers
80

12. POWER AND CONTROL


81
12.1 Power supply
81
12.1.1

Background
81

12.1.2

Design Considerations
81

12.1.3

110 kV / 11 kV Substation within the Plant premises


82

12.1.4

Pumping station electrical works, controls and electrical systems


82
12.1.5

Emergency power generation


83

12.1.6

Electrical Equipment
83

12.2 Control System


84
12.2.1

General Design Criteria


84

12.2.2

Plant Operation Methods


85

12.2.3

Control System Architecture


Page 6

Interim Report

85

13. CIVIL, BUILDING AND STRUCTURAL WORKS


87
13.1 Plant Buildings
87
13.1.1

General
87

13.1.2

Platforms, Access Ways and Stairways


88
13.1.3

Lifting Devices
88

13.2 Road works


88
13.3 Site Services
89
13.4 Storm water Drainage
89
13.5 Fire Systems
89

14. PRODUCT WATER MAIN


90
14.1 Annexure
96
14.1.1

Pipe sizing and pump capacity for 550 MLD


96

14.1.2

Pipe sizing and pump capacity for 2 X 275 MLD


99

14.1.3

Pipe size and pump capacity for 400 MLD


102

14.1.4

Pipe size and pump capacity for 150 MLD


105

15. BENEIFTS OF LONG-TERM O&M CONTRACT


Page 7

Interim Report

108

Executive Summary
Survey Data
Significant findings in respect of Field Work and Historical Data gathering
including the data of existing 100 MLD Neemeli SWRO are as following:
1.

Sea Water quality is poor.


TSS measured in July/August 2013 on the seabed offshore from Perur
( for the proposed 400 MLD SWRO) was, on average, 73.3 mg/L, and 48.7
mg/L at the surface.

The specification for the Nemmeli desalination plant nominates a design


range for TSS of 50 mg/L to 200 mg/L. By comparison, the desalination
plant for Sydney is designed for TSS of 7 mg/L.

Intake water quality data from Nemmeli plant was provided for the period
from and including May 2013 to early December 2013. The data shows
poor water quality on a number of occasions as evidenced by frequent
spikes in turbidity. These spikes in turbidity resulted in plant shutdown on
13 days in the period of record ie a loss of production on 7% of days of
record.
In July 2013 the Nemmeli plant was closed for 5 days; the water quality
data shows that there was no discernible change in salinity in this period.
In the first week of December 2013 the plant was closed for 4 consecutive
days; the water quality data shows that the salinity dropped significantly
over this period indicating the influx of fresh water, which contained high
levels of suspended solids.

There is strong evidence that jelly fish are frequently in abundance along
the Chennai coast and have caused considerable nuisance in power
station intakes.
Page 8

Interim Report

In literature, algal blooms have been documented along both the west and
east coasts of India. It can be expected that algal blooms will impact upon
a desalination plant at Perur from time to time.

3.

Longshore sediment transport is generally to the north between April and


October, and to the south from November to March.

The influence of

littoral drift is significant and the annual net drift takes place in a northerly
direction.

Pre-treatment Processes

High levels of TSS, algae and jelly fish are expected to be drawn in at the
intake to the plant, either constantly or cyclically, and the pre-treatment
plant must be designed to maintain a high quality feed to the desalination
plant, over the long term.

Because of the jelly fish, at the intake offshore a vertical screen with 100
mm c/c have been proposed to have an approach velocity not exceeding
0.1 m/s at peak flows. A fish net is proposed to be placed outside the
vertical mesh screens to reduce the intake of jelly fish.

Fine screens will be provided before the sea water pumps to capture the
jelly fish and minimise the chance of breaking up the jelly fish, which
would make it harder to remove jelly fish particles.

The fine screens will be provided with 3 mm apertures, targeting jelly fish.

.
DAF process may be capable of handling 10-20 mg/L of TSS, but would not
be capable of handling greater than 75 mg/L of TSS.

For this reason

Lamella settlers, preceded by coagulation-flocculation and polymer


dosing have been suggested. It is expected that the Lamella settlers will
significantly reduce the TSS, but will be only partially successful in
removing the larvae, algae and jelly fish particles off neutral buoyancy;
this material should be removed effectively by the following DAF and
Page 9

Interim Report

filtration processes.

A DAF loading rate of 8 m/hr has been nominated, when operating with 20%
recycle. The nominated recycle rate is high, cf 10 to 12 m/hr at other plants,
so as to be able to handle the anticipated high solids loading rate. Lower
recycle rates will be available to the operator when the water quality is good.

The DAF will be followed by Ultrafiltration (UF), due to space constraints. The
gross flux shall be 50LMH.

RO Plant
Membrane Flux and Overall Recovery
The RO plant is a single stage/single pass design with an overall recovery of
46% specified.
SWRO membrane elements of 8-inch diameter have been selected.

Each

pressure vessel will house eight membrane elements. The design treated water
TDS for the RO plant is 300 mg/L. In sizing the RO plant, an average membrane
flux of approximately 13.5 L/m2/hour (LMH) has been selected. The provision
of 8% spare space for more membranes provides the flexibility to reduce the
flux if warranted, to reduce cleaning frequency or reduce power consumption.
Pumping System Design and Pressure Control
Based on preliminary membrane projections for DOW and Hydranautics
membranes, it is anticipated that the required membrane pressure will range
between 51.6 bar and 65.3 bar, depending on feed water and membrane
conditions, and depending on the membrane selected. In order to provide the
above pressure range, and based on the design temperature and salinity
ranges set out depicted in the report, a combination of VSD driven booster
pumps and fixed speed RO HP pumps has been selected to provide operational
flexibility and minimise electrical energy costs.
With this design the RO Booster pumps will provide feed pressure adjustments
Page 10

Interim Report

suitable to cover the required membrane pressures for average to maximum


salinity events. For feed waters with below average salinities further controls
will be required in the form of a HP throttling valve, essentially burning-off
excess energy created by the pumping system upstream.

Potabilisation
This design nominates a target alkalinity of 40 to 60 mg/L as oppose to the ISO
standard of 200 mg/L (Desirable). This is because at this alkalinity, a positive
Langelier Index is achieved, which is the minimum requirement for addressing
the risk of corrosive water. Higher target alkalinity, while desirable (as it would
further reduce the risk of corrosive water by minimising pH change in the final
water) would require additional stabilisation chemicals, and therefore additional
cost.

The nominated target alkalinity thus represents a balanced approach

between water quality and cost.

Product Water Main


All pipes above 1.1 m dia have been proposed as Carbon Steel with internal and
external lining, whereas all pipes below 1.1m are stated to be Ductile Iron with
internal CC lining.

Page 11

Interim Report

1.
1.1

INTRODUCTION
Background

Chennai Metropolitan Water Supply and Sewerage Board (CMWSSB) is the


statutory body responsible for providing water supply and sewerage infrastructure
to the entire city of Chennai, and although its present operation is limited to the
Corporation City limit, the Board is extending its services to the entire Chennai
Metropolitan Area as development progresses.
Water demand for City (for the year 2016)

1133.0 MLD

Industries

37.0 MLD

Bulk water supply to establishments

28.0 MLD

Sriperumpudur

17.7 MLD

SIPCOT Industrial estate

14.0 MLD

Added Areas of CMA


(such as Avadi, Pallavaram, Pammal and Tambaram
etc).
Total Water Demand

329.0 MLD
1559 MLD

Supply

831 MLD

GAP

728 MLD

As a step towards meeting the demand and matching the provision of


infrastructure with the citys growth, CMWSSB has commissioned consultancy
services to conduct prefeasibility studies, along with preparation of DPR, into the
construction of a Sea Water Reverse Osmosis (SWRO) Plant at Neemeli village
along the East Coast Road (ECR), within the premises of existing 100 MLD
Neemeli SWRO plant

1.2

Field Investigations

Page 12

Interim Report

A major component of the seawater quality assessment study is to carry out a


seawater quality monitoring program to characterise seawater for the proposed
seawater intake as well for engineering and process design of the plant.
To fulfil the assignment, data on sea water quality, temperature, salinity, and
seasonal water quality was collected and is presented in sections below.

1.3

Bathymetry

The data collected from the survey reports of existing of 100 MLD Neemeli and the
oceanographic survey carried out for the proposed 400 MLD SWRO to be located
next to Neemeli, the bathymetry chart shows that the depth contours are generally
running parallel to the coast. The seabed exists with the gradient of 1:40-1:70 till 7
m depth and downstream showed the gradient of 1:150-1: 250.

1.4

Scope of this Interim Report


This Report covers draft design, drawing and technical specifications for the
proposed desalination plant and conveyance system to convey potable water to
the city of Chennai. The draft technical specifications are process focussed and
set out the design parameters for the equipment to be supplied for the said plant.

Page 13

Interim Report

2.
2.1

SITE CONDITIONS
Site Appreciation
The proposed site for 150 MLD desalination Plant is located within the premises of
existing 100 MLD Sea Water Desalination Plant at Nemmeli along East Coast Road,
Kancheepuram District, Tamilnadu, India. The site is located at eastern side of East
Coast Road (ECR) at 12 42' 08" North, 80 13' 29" East and is approximately 40 km
of south of Chennai city.
The climatic conditions are characterised by warm dry winters (27 OC average daily
max) and hot wetter summers (39 OC average daily max) with an annual average
rainfall of 1200mm. Cyclones are common in the area and the site is expected to be
effected by approximately 3 per year.

Fig 1:Site Location

Page 14

Interim Report

Fig 2:Site Location with respect to proposed 400 MLD Perur Desalination
Plant
The site for the proposed 150 MLD is spread over an area of around 9 acre,
placed within the premises of existing 100 MLD SWRO Neemeli. The area has
non-uniform width with around 100 m frontage on the sea side, which increases in
the middle of the site towards the ECR, however again on the road frontage is
tapered due to presence of existing Burial ground.
Therefore with the above site conditions, the uniformity of area available is a major
constraint, which in turn condenses the effective space utilization for engineering
and planning of the facility. Therefore with the above physical constraints, the plant
with membrane pre-treatment is proposed for an output capacity of 150 MLD.

Page 15

Interim Report

2.2

Finished Floor Level


The finished floor level of existing Neemeli 100 MLD SWRO plant is 6.5 m, which
had been kept 0.9 m above the last Tsunami level, therefore the Finished Surface
Level (FSL) of RL 6.5 is being proposed for the said plant. Finished Floor levels
shall be set at not lower than RL 6.8 m and equipment plinth levels not lower than
RL 7.0 for ease in operations.

2.3

Climate
Chennai, Tamil Nadu has a tropical wet and dry/ savanna climate (Kppen-Geiger
classification: Aw) with a pronounced dry season in the low-sun months, no cold
season, wet season is in the high-sun months. According to the Holdridge life
zones system of bioclimatic classification Chennai, Tamil Nadu is close to
the tropical dry forest biome.

Figure 3 Temperature profiles

Page 16

Interim Report

Figure 4 Rainfall (monthly averages)


2.4

Typical Oceanographic Conditions


The oceanography of this region is influenced by 3 climatic conditions viz.,
southwest monsoon (June September), northeast monsoon (Mid - October to
Mid - March) and a fair weather period (Mid - March to May). The coast is more
influenced by the northeast monsoon
than other two seasons. Wave action is
high during northeast monsoon and

Fig 5: Typical Cross Section of Coastal


Form

cyclonic period. Storm surge at Nemmeli has been estimated to be up to 4 metres


above chart datum (ie approximately RL 3.5). The coastal current within 5 km
distance from the shore is greatly influenced by wind and tides. The near shore
current

remains

more

dynamic

Fig

and

turbulent due to persistent action of


seasonal wind, high waves and coastal
currents. The distribution of temperature
and salinity indicates that the near shore
water is well mixed without stratification.
The influence of littoral drift is significant

Fig 6: Tidal Varitionas

and the annual net drift takes place in a northerly direction. The near-shore waters
support certain type of demersal fisheries with moderate bottom animal
community. Tourism and beach resorts are developing rapidly in this region.

Page 17

Interim Report

3.

WATER QUALITY

3.1

Seawater Quality

3.1.1

Importance of Sea Water Quality


Knowledge of seawater quality and variability (diurnal and seasonal) is important
for the engineering of SWRO desalination plants. The inherent seawater quality
and composition is a fundamental factor in the entire process technology of a
desalination plant, from pre-treatment selection to process plant design,
remineralisation requirements and impacts on the disposal of brine. These factors
often have a decisive impact on both capital investment costs and operation and
maintenance expenses of the plant.
Temperature and salinity are critical water quality parameters for SWRO design.
Information on the seasonal range in salinity and temperature of the source
seawater and ion composition is required to design a SWRO plant to meet drinking
water quality targets. Salt passage through the reverse osmosis membranes into
the RO permeate, increases with seawater salinity thus having impact on recovery,
membrane age etc. Seawater salinity and temperature also determine the energy
required to overcome the osmotic pressure in the source water with more energy
required to desalinate colder and more saline water.
In the RO process the membranes are subject to both fouling and scaling which
are very sensitive to feedwater quality. Therefore, it is widely recognised in the
desalination industry that the successful long term operation of a SWRO
desalination plant is contingent on the efficiency of its pre-treatment system in
minimising RO membrane fouling and scaling, thus preventing membrane
hydrolysis caused by any oil or petro-hydrocarbons present in the seawater.
Water quality parameters are important for pre-treatment selection and design of a
SWRO desalination plant include, for example, the range in turbidity, total
suspended solids, oil and grease, organics (TOC and DOC) and silt density index
(not often available), nutrients, particle size etc. In addition, it is critical to have an
understanding of what factors impact on seawater quality e.g. marine
hydrodynamics, point and diffuse pollution sources, riverine inputs during monsoon
seasons, marine activities such as shipping, dredging and the occurrence of algal
blooms that may impact on water quality in the vicinity of the sites.
Page 18

Interim Report

3.1.2

Temperature and Salinity Trends


The site, located on the Coromandel Coast on the western side of the Bay of
Bengal, experiences two monsoon periods, the north east monsoon from
October to mid-February when the area receives most of its seasonal rainfall (62%
between mid-October to December) and the south west monsoon from mid-March
to early September (33% of the rainfall). During and following the monsoon
periods, a significant volume of fresh water is flushed into the Bay of Bengal which
may impact on salinity and temperature of seawater at site.

Fig 7: Average Monthly Rainfall in mm, in Chennai

3.1.3

Field Investigations
Water quality data was collected off Perur (Next to Neemeli) in late July - early
August 2013. The data was collected at the surface and at the bottom, on six
transects, at locations at 750 m, 1000 m and 1250 m to sea.

The data collected

has been analysed and the average results for the various analytes are tabulated
below, in Table 1

Page 19

Interim Report

Table 1

Field Investigations - Seawater Quality

Parameter

Unit

Average at Surface

Average at
Bottom

1
2
3
4
5
6
7
8
9
10

Temp. (C)
TDS (mg/l)
Salinity (ppt)
pH
DO (mg/l)
BOD (mg/l)
COD (mg/l)
Colour, Hazen Units
Odour
Taste

28.4
35,100
34.7
8.20
6.61
2.51
25.0
17.7
Odourless
Salty

27.9
35,200
34.7
8.21
5.89
1.83
39.3
18.7
Odourless
Salty

11
12

Total Alkalinity as CaCO3 (mg/l)


Total Hardness as CaCO3 (mg/l)

111.7
6137

116.0
6038

13
14
15
16
17
18
19
20
21
22

Bicarbonate (mg/l)
TSS (mg/l)
Turbidity (NTU)
NO2-N (mg/l)
NO3-N (mg/l)
Total Nitrogen (mg/l)
PO4-P (mol/l)
Total phosphorus (mol/l)
NH3-N (mol/l)
Magnesium as Mg (mg/l)

127.8
48.7
1.10
0.45
2.68
11.8
0.76
1.45
0.20
1191

126.3
73.3
1.88
0.97
4.08
14.5
0.93
1.36
0.18
1258

23
24
25
26
27
28
29
30
31
32
33

Calcium as Ca (mg/l)
Chloride as Cl (mg/l)
Potassium as K (mg/l)
Sulphate as SO4 (mg/l)
Sulphide as H2S (mg/l)
Colloidal Silica (mg/l)
Reactive Silica as SiO2 (mg/l)
Free Residual Chlorine (mg/l)
Fluoride as F (mg/l)
Oil & Grease (mg/l)
Phenolic compounds as C6H5OH

367
18302
375
2758
<0.01
<0.02
0.84
<0.2
1.49
<2
<0.01

467
18502
391
2878
<0.01
<0.02
1.38
<0.2
1.63
<2
<0.01

<0.5
3.25

<0.5
3.17

(mg/l)
34
35

Barium as Ba (mg/l)
Boron as B (mg/l)

Page 20

Interim Report

36
37
38
39
40
41
42
43

Cadmium as Cd (mg/l)
Cyanide as Cn (mg/l)
Iron as Fe (mg/l)
Lead as Pb (mg/l)
Manganese as Mn (mg/l)
Selenium as Se (mg/l)
Total Arsenic as As (mg/l)
Zinc as Zn (mg/l)

<0.01
<0.05
<0.01
<0.01
<0.05
0.022
<0.01
0.092

<0.01
<0.05
<0.01
<0.01
<0.05
0.023
<0.01
0.088

< indicates below the detection limit.


Nemmeli 2013: Intake water quality data from Nemmeli plant was taken for the
period from and including May 2013 to early December 2013, see Figure below.
The data shows poor water quality on a number of occasions as evidenced by
frequent spikes in turbidity. These spikes in turbidity resulted in plant shutdown on
13 days in the period of record ie a loss of production on 7% of days of record.
In July 2013 the Nemmeli plant was closed for 5 days; the water quality data shows
that there was no discernible change in salinity in this period. In the first week of
December 2013 the plant was closed for 4 consecutive days; the water quality data
shows that the salinity dropped significantly over this period indicating the influx of
fresh water, which contained high levels of suspended solids.
.

Page 21

Interim Report

3.1.4

Water Quality Design Criteria

The sea water quality adopted for design purposes is tabulated in and
is based on the water quality sampling program as cited above

Table 3 and

In the Normal values are based on the average data gathered from the field. The
Minimal and Maximum adopted values draw upon data gathered from the area, as
set out above, and upon the experience of the designers.

Page 22

Interim Report

Table 2

Seawater Quality Design Criteria

Criteria Description

Water Temperatures
- surface
- bottom
Turbidity
Total Suspended Solids
pH
Total Dissolved Solids

Unit

C
C
NTU
mg/L
mg/L

Normal based
on field
measurements

Minimum adopted

Maximum adopted

28.4
27.9*
<10
75
8.2
35,200

26.0
25.0
10
50
7.7
32,000

30.0
32.0*
125
200
8.5
38,000

Note: * The design of the RO Plant will make provision for a rise in seawater temperature of 1 oC
resulting from the energy input at the pumps.

In

Table 3, the design values for the proposed plant are stated herewith.
Table 3

Seawater Quality Design Criteria Dissolved Solids

Criteria Description

Unit

Normal based on
average field
measurements

Temperature
Total Dissolved Solids
Bicarbonate (HCO3)
Magnesium as Mg
Calcium as Ca
Chloride as Cl
Potassium as K
Sulphate as SO4
Reactive Silica as SiO2
Fluoride as F
Barium as Ba
Boron as B
Sodium Na
Nitrate NO3
Ammonium

C
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L
mg/L

27.9
35,200
126.3
1258
467
19,247
391
2,878
1.38
1.63
0
3.17
10,789
4
0.2

Page 23

Interim Report

3.2

Water Quality Targets


CMWSSB has specified that the water quality targets shall be as per IS 10 500:
1991 (reaffirmed in 1993). The quality parameters as per IS 10 500 1991 are
indicated in Table 4. It is to be noted that the water quality requirements fall into
two groups, Essential Characteristics and Desirable Characteristics.
Table 4: Drinking Water Quality Requirements as per IS 10 500 1991
Sl. No

Substance or characteristic

Requirement

Essential Characteristics

(Desirable

i.

Color, Hazen units, Max.

ii.

Odour

Unobjectiona

iii.

Taste

ble
Agreeable

iv.

Turbidity, NTU, Max.

v.

pH value

6.5 to 8.5

vi.

Total Hardness (as CaCO3) mg/l, Max.

300

vii.

Iron (as Fe) mg/l, Max.

0.3

viii.

Chlorides (as Cl) mg/l, Max.

250

ix.

Residual, free chlorine ,mg/l, min.

0.2

x.

Fluoride (as F) mg/l, max.

1.5

Desirable Characteristics
xi.

Dissolved solids mg/l, Max.

500

xii

Calcium (as Ca) mg/l, Max.

75

xiii.

Magnesium (as Mg) mg/l, Max.

30

xiv.

Copper (as Cu) mg/l, Max.

0.05

xv.

Manganese (as Mn) mg/l, Max.

0.1

xvi.

Sulphate (as SO4) mg/l, Max

200

xvii.

Nitrate (as NO2) mg/l, Max.

45

xviii.

Phenolic compounds (as C6H5OH) mg/l, Max

0.001

xix.

Mercury (as Hg) mg/l, Max

0.001

xx.

Cadmium (as Cd) mg/l, Max

0.01

xxi.

Selenium (as Se) mg/l, Max

0.01

xxii.

Arsenic (as As) mg/l, Max

0.01

xxiii.

Cyanide (as CN) mg/l, Max

0.05

xxiv.

Lead (as Pb) mg/l, Max

0.05
Page 24

Interim Report

xxv.

Zinc (as Zn) mg/l, Max

xxvi.

Anionic detergents (as MBAS) mg/l, max

0.2

xxvii.

Polynuclear aromatic hydrocarbons

Chromium (as 0.05

Cr6+) mg/l, Max


xxviii.

(as PAH) g/l, Max

xxix.

Mineral oil mg/l, Max

0.01

xxx.

Pesticides mg/l, Max

Absent

Substance or characteristic
xxxi

Radioactive materials *
a) Alpha emitters Bq/l, Max

0.1*

b) Beta emitters pci/l, Max

1*

xxxi.

Alkalinity mg/l, Max

200

xxxii.

Aluminium (as Al) mg/l, Max

0.03

xxxiii.

Boron mg/l, Max

xxxiv

Langlier Saturation Index

positive

3.2.1

Total Dissolved Solids


IS 10 500 states that one of the desirable characteristics of drinking water is that
the maximum dissolved solids be limited, and not greater than 500 mg/L. For the
purposes of this report the design target has been set at 300 mg/L for permeate,
such that the dissolved solids in drinking water, after potabilisation, will not exceed
500 mg/L.

3.2.2

Boron
Boron emerged as a water quality issue with the increasing use of desalinated
seawater for both municipal drinking water supply purposes and for irrigation. The
human health effects of boron have been debated for some time with increasing
confidence, and higher allowable concentrations, from studies over the last 10
years. This is reflected by increases in guideline standards, whereby WHO have
changed their guidelines over the last ten years from 0.3, to 0.5 (2008) and now to
2.4 mg boron/L in the most recent (fourth) edition of the guidelines. A summary
statement on the World Health Organisations (WHO) recommended boron
Page 25

Interim Report

guideline value has been posted on the WHO Water Sanitation and Health website
at:

www.who.int/water_sanitation_health/dwq/chemicals/boron/en/.

The

new

guideline value of 2.4 mg/L was based on a human health point of view. The level
was derived based on an assumption that 40 per cent of a 60 kg adults total daily
intake of boron is a result of drinking 2 litre of water having a boron concentration
of 2.4 mg/L and a safety factor of 60x.
While sodium and chloride ions are well rejected by RO membranes, boron is not
well rejected because a high portion of the boron species is uncharged. The
rejection of boron is strongly dependent on feedwater pH. Under optimal neutral to
alkaline conditions, boron rejection is 70 to 80%, reducing to as low as 50 to 60%
under more acidic feed conditions.
IS 10 500 states that one of the desirable characteristics of drinking water is that
the maximum concentration of boron be not greater than 1 mg/L.

Modelling

indicates that single pass RO can reduce the concentration of boron to between
0.4 and 1.0 mg/L, assuming a maximum feedwater concentration in the source
water of 3.4 mg/L, depending on the age and type/brand of the membranes.
For the purposes of this report the design target for boron has been set at 0.75
mg/L at commissioning, ie with new membranes. The concentration of boron in
the permeate is expected to rise as the membranes approach their AMLT, and it is
expected that minor variations to the process operation, primarily adjustment of the
pH and the addition of anti-scalant, will allow the concentration of boron to be
maintained below the long term goal of < 1.0 mg/L.

3.2.3

Jelly Fish Blooms


Jelly Fish
Jelly fish ingress in a power station cooling system is reported on by S. Rajagopal
et al, Home 1989. The power station is located at Kalpakkam on the east cost of
India. Three peaks in the occurrence of jelly fish were noticed in May, July and
October 1988, which accounted for 286, 204 and 129 tonnes respectively. The
authors recommend that jelly fish ingress be controlled by preventing their inflow,
Page 26

Interim Report

or by letting them pass freely. To prevent jelly fish from flowing in with water
current, fish netting is proposed to be installed in front of the water intake where
inlet velocity is low. This method is adopted at many power stations including
Tarapur Atomic Power Station including existing SWRO plants.

4.

DESIGN

LIFE

AND

AVAILABILITY;

PRIMARY

FLOWS
4.1

Design Life

The infrastructure will be designed for a life in excess of:


Page 27

Interim Report

4.2

50 years for civil works, buildings & buried pipeline components;

50 years for concrete tanks.

25 years for mechanical and electrical equipment;

20 years for flexible membrane liners and covers to storages;

20 years for steel panel tanks;

15 years instruments;

10 years for polyethylene tanks;

8 years seawater reverse osmosis (SWRO) membrane life.

Operations and Availability


In order to specify design parameters for individual processes, the proposed
operation and availability of the RO plant is for foremost importance. Therefore all
the design parameters are based on 95% plant availability throughout the year.

4.2.1

RO Rack Configuration
The size of the RO train impacts on the number of trains, all equipments as well
as the pipework required, along with pump set efficiencies. To have the most
optimal of above, the RO racks have been selected within each train having the
following characteristic:

Table 5 : RO Racks (150 MLD Module)


Design Parameter

Details

RO Configuration

Single pass, single stage

Number of Duty Trains

Number of Rows per Train

13

Number of Columns per Train

20

Number of slots available

260

Number of PVs per Train required

240

Spare space for additional PVs

8% (20 PVs)

Number of Membranes per PV

Number of Membranes per Train

1920
Page 28

Interim Report

As shown in Table 5 space has been provided in the RO racks for additional PVs,
which may be installed to respond to operational and environmental changes, as
well as to compensate for deviations from the design of the plant or equipment
performances. The additional PVs would not be installed to increase the capacity
of the trains above the design value, but should be used to ensure maintenance of
the design capacity of the trains.
The RO building will be designed to house six trains, in a single row. There shall
be no standby train.
4.3

Primary Flows
Based on the proposed operation and availability of the plant as described above,
primary flows have been calculated and are tabulated below in Table 6. It is to be
noted that these flows are not the final design flows and can be expected to vary
slightly as the designs are developed.
Table 6 Primary Flows 150 MLD Plant
Section
Product Water Tank to Chennai

150 MLD

Utilities and Losses

3 MLD

Permeate Production

153 MLD (6 trains)

Permeate per Train

25.5 MLD

Peak Filtered Seawater Required to 356 MLD


RO (based on 43% recovery *)
Filtered Sea Water to Screens for 1.3
spray
UF Losses

30.0 MLD

Sludge and Float losses

7.0 MLD

To Pre-treatment

394.30 MLD say 395 MLD

Screened

Sea

Water

to

Pre- 395 MLD

treatment
Intake Flow

394 MLD

Peak Outfall Rate

242 MLD

Page 29

Interim Report

43% selected to provide a conservative estimate of flow for hydraulic design.


Process design is based on 46% recovery.

5.
5.1

SEAWATER INTAKE
Primary Flows
The intake system will be designed to have a hydraulic capacity of 394 MLD,
whereas the sea water pumping station will be designed to have a peak capacity of
395 MLD, wherein 1.0 MLD required for screen washing will be returned upstream
of screen.
The intake conduits shall be sized with 10% additional capacity to take care for
bio-growth which shall result in reduction of carrying capacity. Therefore the inlet
pipe shall be sized for a flow of 435 MLD.
The intake conduit shall be of 2100 mm (ID) diameter and the outlet conduit shall
Page 30

Interim Report

be of 1400 mm ID diameter to produce 150 MLD permeate capacity. The velocity


in the conduit at peak flow will be 1.45 m/s. Friction losses at this velocity would
be of the order of 1.5-1.6 metre in a pipeline 950 metres long.
One outfall will be provided with an internal diameter of 1400 mm. For a flow peak
flow of 242 MLD, the velocity will be 1.82 m/s. Friction losses at this velocity would
be of the order of 1.8-2.0 metre in a pipeline 650 metres long.
The exit losses through multiple ports at 7 m/s would be of the order of 2.5 metres,
giving a total head loss in the outfall of around 4.7 metres say 5 meters at peak
flows.
As cited in section above, the site frontage along sea side is nearly 100 m and
there are existing intake and outfall pipes, therefore it is proposed that the intake
and outfall be south of existing plant intake, with proposed outfall north of
proposed intake.

5.2

Intake structure with Screen offshore


A 100 mm screen in GRP construction will be provided at the intake to exclude
larger marine life. The screens will be 7 m diameter, 1.5 m high, starting 2 m
above the sea floor, in 9 metre depth of water. The approach velocity will be <0.15
m/s to minimise the entrapment of marine species.

A fish net with will be provided to minimise the ingress of jelly fish to the intake.
The fish net will be required to be inspected and replaced from time to time, as the
same is likely to be damaged by marine lives. The head loss through the intake
system will also be monitored, and in any increase in system losses indicating
fouling at the intake, or the growth of biomass within the intake conduit, the same
shall be cleaned through divers.

5.3

Active Screens before the Sea Water Pumps


Travelling Band Screens shall be provided before the Pumps on shore and shall
be the first level of defence to the plant against all foreign matter like floating, sea
shells, diatoms etc.
Page 31

Interim Report

The advantage of providing the screens before the pumps is to capture the jelly
fish that bypass the off shore screen and fishing net on the shore alive, thus
preserving the same and returning them to the sea in order to have a sustainable
ecosystem.
A typical band screen installation is shown in Figure 8 below.

Band screens have been in service for screening sea water for many years. They
provide efficient removal with relatively low maintenance costs and also have
Construction costs that are relatively low as compared to the total plant cost.
Through-flow band screens have been proposed herewith. Typically mesh sizes
vary from 2 mm to 10 mm, and, in view of the marine biomass problems at
Nemmeli, a mesh size of 3 mm has been selected.
Table 7 : Band Screen Design Parameters
Parameter

Units

Value

Seawater flow

MLD

395
Page 32

Interim Report

5.4

Duty

Stand by

Sea Water Flow Per screen

MLD

198

Stainless Steel mesh

mm

Approach Channel Width

4.0

Width Back opening

2.8-3 m

Channel Approach Velocity

m/s

0.85-0.90

Upstream water depth

2.2

Downstream water depth

1.8

Wash Water per screen

L/s

15-20

Sea Water Pumping Station


There shall be one number of 2,100 mm internal diameter intake conduit. The
conduit shall be constructed by pipe jacking from an onshore pit, which will
become/converted into the sea water pumping station once the conduits have
been installed.
Vertical shaft pumps in a wet well are particularly suitable for sites with a low tidal
range, such as is experienced at Chennai. Vertical shaft pumps in a wet well have
been selected.

The tides at Perur, next to the Nemeeli as well as site ( data taken from DPR of
Existing 100 MLD SWRO Neemeli) are as follows, in metres above Chart Datum,
and relative to the site datum:

Chart Datum

RL

Mean High Water Spring

1.15

RL 0.5

Mean High Water Neaps

0.84

RL 0.19

Mean Sea Level

0.65

RL 0.0

Page 33

Interim Report

Mean Low Water Neaps

0.43

RL 0.22

Mean Low Water Spring

0.14

RL 0.51

Onshore survey levels are recorded as metres above Mean Sea Level. Thus
Mean High Water Springs is 1.15 0.65 = RL 0.5, and Mean Low Water Springs is
0.14 0.65 = RL 0.51.
The low no flow level in the pump well has been taken as RL 0.50. At a flow of
395 MLD in intake conduit, the hydraulic grade line entering the proposed pumping
station will be approximately 1.7 m lower, at RL 2.2. A head loss of 0.8 is
provided/allowed for loss through screens resulting in low-low water levels in sump
cum pump station as RL-3.0.
Three (3) metres has been provided to the bottom of the pumps, which has been
assumed to be 0.5 m above the bottom of the pumping station at RL 6.5 to
ensure sufficient submergence for the pumps, and to allow for surging in the intake
line. Subject to the detailed design of the intake conduits, it is anticipated that the
intake conduit will enter the pumping station at between RL 6.5 and RL 4.5.
Therefore, the static lift at Mean Low Water Springs will be from RL -3.0 to around
RL 14.5, ie around 17.5 metres; and the static lift at Mean High Water Springs will
be from RL -2.0 to around RL 14.5, i.e. around 16.5 metres.

Pump Configuration
There shall be 3 W + 2 S pumps sets in wet well delivering as cited below
Three working pumps, delivering to a common manifold of 2.0 m diameter, with
a two pump as standby
With 3 pumps in operation each pump must be capable of delivering 132 MLD
(1528 L/s), at a head of around 25 metres.
Each pump will have a delivery line 1.1 metres in diameter (V = 1.60 m/s),
connected to the manifold 2.0 metres in diameter.
Page 34

Interim Report

Isolation valves along with flow meters shall be installed main and branch
manifolds.
These pumps are Vertical Turbine pumps with large clear passages. The intake
screen and net at the intake will prevent the ingress of material like to cause a
blockage.

Design parameters are set out in Table 8


Table 8 :Sea Water Intake Pumps

Parameter

Units

Value

Pumping Station Capacity

MLD

395

Wet Wells

Duty Pumps

Standby

Mean High Water Springs

RL

0.5

Mean Low Water Springs

RL

- 0.51

Intake Conduit loses at peak flow

1.7-1.8

Pipe diameter - Pump to manifold

1.1

Main Rising main diameter

2.0

Pump house Plinth Level

RL

7.0

Generic type of pump: vertical shaft mounted in pump well,

Materials: super duplex stainless steel, PERN No > 43

Equipment per pump unit: isolation valves, non-return valve and magnetic flow
meter,

5.5

Pump motors: electric variable speed,

Pump station control: by plant PLC based upon target flow rate,

Pump/equipment removal: by EOT,

Shock chlorination system


A shock chlorination system is proposed to minimise marine growth in the inlet
Page 35

Interim Report

pump station and pressure main. Criteria adopted for the shock chlorination
system are summarised below;

5.6

Type of chemical: chlorine gas through Vacuum Chlorinate

Max chlorine solution residual in inlet main: 8 mg/l,

Min chlorine solution residual in inlet main: 1 mg/l,

Dosing time: 2 hour per day,

Dosing point: Off shore inlet well,

For the same 45 kg/hr (3 W + 1S) Vacuum Chlorination is provided

Intake / Outfall Pipes


Jacking pipes shall be of precast concrete, designed and manufactured to meet
the requirements of:

BS5911 Specification for reinforced concrete jacking pipes with flexible joints.

BS EN ISO 9002 Manufacturing.

Or equivalent standards as appropriate.


All jacking pipes, including special designs, shall be sufficiently reinforced with
steel to withstand all stresses induced by handling, jacking, earth and water
pressures and all working loads at the depths at which they are to be used without
cracking, spalling or distortion. A load factor of not less than 1.5 shall be used in
the calculations to determine the strength of the pipes required. The clear cover to
any steel reinforcement shall not be less than 50mm.
The strength of the pipes shall be tested by a three edge bearing test. When
subjected to the design load in such a test, the maximum crack developed on the
pipe shall not exceed 0.25mm.
All pipes shall be manufactured by a supplier approved by the engineer. The
process of manufacture may be by centrifugal or vertical casting subject to
submission and approval by the Engineer.
The steel collar shall be fixed to the reinforcement cage prior to casting of the pipe,
to ensure full integration of the pipe components.
Pipes shall be fully cured and have reached the designed concrete strength prior
Page 36

Interim Report

to delivery.
5.6.1

Pipe jacking
The tunnel shall be constructed by pipe jacking techniques using a purpose made
TBM for excavation. The TBM shall be designed as a slurry machine offering
controlled support pressure to the tunnel face and allowing for the use of bentonite
Slurry where required. The complete system shall be purpose designed for the
project and shall include TBM, slurry, circuit, main jacking station, interjack stations
and automatic bentonite injection system for lubrication of the pipe annulus.
The jacking pipes for this project shall be 2100mm and 1400 mm I/D. They shall
be using steel (Super Duplex : PERN>43) banded joints with full section contact
between adjacent pipes. The design of the pipes shall be sufficient to withstand
the jacking and overburden pressures assuming that inter jack stations are
installed at 100m spacing. The design should also allow for driving the tunnel at a
100-150m radius.

5.6.2

Main Jacking Station


The main jacking station shall be of robust construction, and comprise a thrust ring
and cylinders fixed firmly to a back wall and slide rails.
The thrust ring only shall be used for advancing pipes. A local control shall be
provided at the pit bottom for use by the shaft bottom crew during pipe extensions
only. During normal advance the main jacking station is controlled from the TBM
operators console, however Cylinder pressures should be available to the operator
at all times.
An interlock arrangement shall exist to transfer control between the operators
console and the shaft, whereby only the operator in ownership of the control can
make the transfer at the time.

5.6.3

Inter-jacks
Interjack stations shall be designed to provide sufficient jacking force to overcome
the frictional resistance to pipe motion over the distance between them. The
interjack stations should have a minimum capacity of 1200 tonnes and a maximum
Page 37

Interim Report

spacing of 300 m. The seal shall be adjustable to exclude the ingress of ground
water and bentonite and shall be capable of withstanding working pressures in
excess of 4bar.
Hydraulic power to operate the inter jack stations shall be provided by power packs
situated in both the TBM and in the launch shaft. In order to minimise hose lengths
and subsequent power losses, the forward inter jacks shall be fed from the TBM
power pack and the rear inter jacks, from the shaft power pack.
On completion of tunnelling operations all interjack cylinders and seal assemblies
shall be removed. The exposed steelwork shall be properly cleaned and treated
with epoxy paint suitable for sea water conditions and the pipes shall be closed to
eliminate the gap. The adjustable interjack sealing systems shall be manufactured
to facilitate easy removal by hand within the tunnel.
5.6.4

Slurry System
The slurry system shall be designed to be capable of functioning with bentonite
slurry where required as a support medium. The slurry circuit shall be designed for
a minimum flow rate of 400m3/hr. The slurry circuit shall include flow meters on
both feed and discharge lines.
Feed pipes shall be 250mm nominal internal diameter and slurry discharge 200mm
internal diameter. Automatic valves shall be included at the shaft top to allow for
immediate closure by the operator in the event of burst system hoses. The
separation plant shall be sized to be 25% over capacity to account for any possible
flow surges within the slurry circuit.
All controls and indicators for running the slurry system shall be available at TBM
operators console.

A telephone system shall be installed to allow direct

communication between the TBM and the separation plant.

6.
6.1

PRETREATMENT SYSTEM
Pretreatment Process
The pre-treatment process has been sized assuming that there are 6 membrane
trains in operation to produce 150 MLD permeate.

Page 38

Interim Report

6.2

Sea Water Quality / Treatment Processes


The key water quality parameters relevant to the selection of pre-treatment
processes include:

Total Suspended Solids

Turbidity

Marine Biomass, jelly fish, larvae and algae

Other relevant factors are:

Capex

Opex

Ease of operation

Robustness

6.2.1

Key Water Quality Parameters Pre-Treatment


Field Investigations
The results of Field Investigations into sea water quality are tabulated in Table 1.
The key water quality parameters in respect of the design of pre-treatment facilities
are discussed below.
Turbidity
The average of the turbidity measurements at the surface and at depth, offshore
from site, were 1.10 and 1.88 NTU respectively. These readings are normal for
sea water. Notwithstanding, turbidity must be reduced to less than 0.5 NTU to
protect the RO membranes.

Total Suspended Solids


The average of the TSS measurements at the surface and at depth, offshore from
site, was 48.7 and 73.3 mg/L respectively. These readings are very high. For
example, for Sydney desalination plant (250 MLD) the design maximum TSS is 7
mg/L.
The higher concentration of suspended solids at the bottom compared to the
Page 39

Interim Report

surface indicates that the suspended solids have a density slightly higher than sea
water.

Since the seawater intake is located near the seabed, the bottom

suspended solids readings are quite relevant to the design of the pre-treatment
system.

Marine Biomass
Marine biomass includes jelly fish, larvae and algae.

The presence of these

materials is highly variable and dependent on season, temperature and the


presence of nutrients.
However, operational experience at Nemmeli suggests that if provision were not
made for the removal of marine biomass it could be a significant problem; hence to
take care of same a Travelling Band Screen with 3mm mesh has been provided.

Organic Pollutants
The presence of organic pollutants is indicated by measurements such as BOD,
COD, NO2, NO3, and NH3. BOD is relatively low, however, COD is much higher;
the reason for high COD is not clearly correlated, but may be related to the high
suspended solids.

6.2.2

The Pre-Treatment Processes- Scheme

pH correction coagulation flocculation Lamella Settlers- Dissolved Air


Flotation (DAF) Ultra Filtration(UF)

Page 40

Interim Report

6.3

pH Correction
The seawater will be dosed with acid to achieve the optimum pH for coagulation.
The dosing system will consist of duty and standby dosing lances into each pretreatment train. The pH correction design parameters are summarised in

Table

9 below.
Table 9

pH Correction Design Parameters

Parameter

Units

Value

Design maximum seawater flow

MLD

395

Dosing location

At

head

of

Common

Coagulation/Flocculation
Type of pH correction chemical

Sulphuric Acid

Target pH

6.80

Dose rate
Number of dosing pumps
6.4

mg/L

35 approx.
2 duty / 1 standby lances

Coagulation
Seawater turbidity is usually low around 0.5 to 1.0 NTU, however at the proposed
site is high. Turbidity normally indicates the presence of mineral particles which
can cause colloidal-particle fouling of RO membranes, thus reducing the flux rate
of the membranes. For this reason it is important to reduce the turbidity to an
acceptable level, i.e. to less than 0.5 NTU, perhaps as low as 0.1 NTU.

A widely used test is the Silt Density Index (SDI). The test is a laboratory filtration
test detailed in an ASTM standard.

The aim of the pre-treatment system is to

produce filtered seawater with a Silt Density Index (15 minute test as per ASTM
D4189 95-2002) of less than 3.0 for 95%tile and less than 3.5 for 100%tile of the
time, under all operating conditions, to produce 150 MLD of permeate.

Mineral particles can be destabilized by the addition of coagulant. The pH is


adjusted first, then the coagulant is added and flash mixed with the water.
Coagulation is followed by flocculation; a flocculant is added at this stage.
Page 41

Interim Report

Thereafter the flocs are removed by processes such as floatation and filtration.
Coagulant will be dosed at the head of Lamella Settler block. Rapid mixing will be
implemented within a concrete structure with at least 30 seconds of detention time.
The coagulation and rapid mixing design parameters are summarised in Table 10
below.
Table 10: Coagulation and Rapid Mixing Design Parameters

6.5

Parameter

Units

Value

Design maximum seawater flow

MLD

395

Type of coagulant

Ferric chloride

Mixer type

Flash mixer

Mixing energy (G)

s-1

600 minimum

Mixing time

30

Flocculation
Flocculation will be provided again at the head of each Lamella Settler.
Flocculation will be achieved in two flocculation tanks in series with each tank
consisting of one or two vertical mixers so as to have the mixing energy tapered.
Flocculent (flocculation aid) will be dosed at the entry to each flocculation tank.
Due to the warm temperature, 20 minutes of flocculation is proposed. The longer
flocculation time is proposed so as to compensate for the lower suspended solids
concentration expected as feed to RO. The flocculation and flocculent design
parameters are summarised in Table 11 below.
Table 11 Flocculation and Flocculation Aid Design Parameters
Parameter

Units

Value

Design maximum seawater flow

MLD

395

Type of flocculation aid

polyelectrolyte

Dose rate average

mg/L

0.1

Dose rate maximum

mg/L

0.5

Number of flocculation tanks per cell

No.

Detention time (minimum)

min

20
Page 42

Interim Report

Type of mixers

6.6

Variable speed impellers

Mixing energy (G) stage 1

s-1

70 100

Mixing energy (G) stage 2

s-1

30 - 50

Lamella Settlers
Lamella settlers consist of a series of inclined parallel plates or tubes through
which the water passes.

The

plates are close together and the


suspended solids travel only a
short distance to the plate below
and then gravitate down the
inclined plate and are removed
from the flow. For this reason the

footprint of a Lamella settler is relatively small. Lamella settlers are considered


excellent roughing filters, ie Lamella filters can quickly remove large amount of
relatively coarse material. However, the turbidity of the water exiting a Lamella
settler may be higher than for plain sedimentation. Notwithstanding, low turbidities
may be achieved by providing filters following the Lamella plate settlers.
Coagulation and flocculation are provided prior to the Lamella settlers. The capex
and opex of Lamella plate settlers are reasonably low, and they are simple to
operate.
Table 12 Lamella Settlers Design Parameters
Parameter

Units

Value

Design maximum seawater flow

MLD

395
Page 43

Interim Report

No. of settlers

Nos.

Lamella Type and Material

8
Plates/Tubes made from FRP
with 550-600 inclination with
horizontal

Lamella

Surface

loading

rate m/hr

<15

(vertically projected surface area)


Lamella Supports
Bottom Sludge Removal

Concrete
Chain and flight scrapers or
hydraulic vacuum; material of
construction to be suitable for

Clarified water collection

seawater
Concrete

troughs

with

triangular FRP weir plates


6.7

Dissolved Air Flotation (DAF)


Dissolved air floatation (DAF) is a high rate process using micro-bubbles to float
coagulated and flocculated particles to the surface of a clarification basin.

portion of the clarified water is drawn off and passed through a pressure vessel
where the water is saturated with air at high pressure. The pressurized water,
supersaturated with air, is released into the water as it enters the bottom of the
basin. The sudden release of pressure causes the formation of micro-bubbles
which attach to the preformed floc and carry it to the surface where it forms a thick
floating layer. The float is skimmed off, and the clarified water passes to filtration.
DAF in general adequately removes TSS of 75 mg/L, but above 100 mg/L there is
likely to be a higher carryover of solids to the filters, resulting in shorter run times.
The potential for short filter runs can be offset to some extent by designing the
filtration system at an optimal flux. Therefore to mitigate the said situation; the
membrane filtration has been designed at an optimal flux of 50 LMH, while
producing the required SDI.
DAF would be effective in the removal of algae, and should be reasonably effective
in the removal of other marine biomass, such as larvae and jelly fish particulate
matter. DAF would be effective in the removal of oils and greases.
Ideally, loading rates for DAF would be set after pilot plant testing, and if time
Page 44

Interim Report

permits, pilot plant testing at site, should be carried out to provide data on the most
suitable loading rate.
DAF a form of physical treatment of water that involves coagulating impurities then
removing these solids by using dissolved air to float them to the surface of the
water. Residual fine particles that are not floated off are removed by the filtration
downstream.
The flotation cell also requires occasional sludge removal to remove settled solids.
The DAF process is proven across the globe for a range of water treatment
applications including as a pre-treatment process in sea water desalination plants
for removal of fine solids and organics. It also copes well with algae because it is
removed in the flotation step, eliminating the possibility of filter blinding. A
simplified cross sectional schematic is as shown below.
Raw
water

Flotation
Flocculation
Filtration

Air
saturated water

Sludge

Filtered water

Table 123

Dissolved Air Flotation Design Parameters

Parameter

Units

Value

Design seawater flow

MLD

395 without recycle

MLD

474 with recycle

Number of DAF Cells

No.

22

DAF surface loading All units online

m/hr

8 (with recycle)

m/hr

8.5 (with recycle)

DAF area per cell

m2

95

DAF Recycle Rate

20

DAF Air Dose Rate

mg/L

10

rate
One unit offline

Page 45

Interim Report

Parameter

Units

Value

Air Saturator Efficiency

90

Air Saturator Pressure (Gauge)

kPa

800 (nominal)

Number of air saturator

No.

22

Saturator volume

7,500 (at 90 s detention time 50% full at 20% recycle)

Float Removal Type

Hydraulic float removal (and weir)

DAF recycle water will be drawn from the RO fed tank i.e. UF filtered water.
6.8

Ultrafiltration (UF)
Membrane filtration involves passing the water to be treated through a thin polymer
film called the membrane. The filtration process is loosely referred to as
microfiltration if the size of pores in the membrane is about 0.2 micron.
If the pores are smaller than approximately 0.1 micron the filtration process is
referred to as ultra filtration.
One of the key advantages of membrane filtration is that due to the small pore size
compared to media filtration, membrane filters usually require less coagulant to
achieve the required filtered water quality. Further since being a pressurized
membrane system the area required is generally 25-30% less than non-membrane
filtration.
The

pressure

arrangement

is

similar to an RO
plant

with

ultra-

filtration membranes
installed in pressure
vessels.

Feed

water is pumped through the membranes and unlike RO, UF is vertical in


configuration. However, with recent development even horizontal configurations are

Page 46

Interim Report

possible in UF like RO. Pressure systems are typically supplied as modular skid
mounted systems with skids varying in size depending on the duty requirement.
Table 14

Ultra Filtration Design Parameters

Flow to Self-

MLD

388

99.5%

MLD

386

92.5%

Flux

LMH

50

Total Number of

Nos.

32

Duty

Nos.

30

Standby

Nos.

Flow Rate to RO

MLD

356

Area of each skid

m2

17.5 x 4.5 m

Initial TMP

Bar

0.5

Cleansing filters
Assumed recovery from filters
Flow inlet to UF
UF Recovery Rate

skids

6.9

Ultrafiltration Water Storage


Filtered water i.e. UF permeate storage will be located immediately downstream of
the UF unit. This will serve as feed tank to Reverse Osmosis.
The tank has been sized to provide twenty minutes storage at a seawater flow of
388 MLD.
The backwash water for UF shall also be withdrawn from the same tank.
The design parameters for the Filtered Seawater Storage Tanks are summarised in
Table below.

Filtered Seawater Storage Design Parameters


Parameter

Units

Value

Design seawater flow

MLD

388
Page 47

Interim Report

Number of tanks

No.

Fill time

min

20

Capacity

ML/tank

5.5

Page 48

Interim Report

7.

REVERSE OSMOSIS PLANT

7.1

Overview of RO Plant
The RO plant will be a single stage/single pass design with an overall recovery of
46%.
The plant will be configured as six numbers of individual trains. Each of the six RO
trains will consist of one RO rack each, with dedicated pumping system and
Energy Recover Devices (ERDs).

The plant is to be designed such that full

production can be achieved through six trains. The trains will be configured to
allow for each individual train to be isolated for cleaning, maintenance, or
membrane replacement. However 2 sets of standby ERD, HPP and booster
pumps have been proposed.
SWRO membrane elements of 8-inch diameter have been selected.

Each

pressure vessel will house eight membrane elements.


The design treated water TDS for the RO plant is 300 mg/L. No allowance has
been made in the RO plant layout for a second pass RO system.
In sizing the RO plant an average membrane flux of approximately 13.5 L/m2/hour
(LMH) has been selected. The provision of 8% spare space for more membranes
provides the flexibility to reduce the flux if advantageous for reducing cleaning
frequency or reducing power consumption.
7.2

Pumping System Design and Pressure Control


Based on preliminary membrane projections for DOW and Hydranautics
membranes, it is anticipated that the required membrane pressure will range
between 51.6 bar and 65.3 bar, depending on feed water, the membrane
conditions, and depending on the actual membrane selected.
In order to provide the above pressure range, and based on the design
temperature and salinity ranges as depicted in Section 3, a combination of VSD
driven booster pumps and fixed speed RO HP pumps has been selected.

schematic of the proposed pumping system design is shown Figure below.

Page 49

Interim Report

Under such design the RO Booster pumps would provide feed pressure
adjustments suitable to cover the required membrane pressures for average to
maximum salinity events. For feed waters with below average salinities further
controls would be required in form of a HP throttling valve, essentially reducing
(burning-off) excess energy created by the pumping system upstream. It should
be noted that correct sizing of the RO pumping system is critical, as under-sizing
will not achieve the required permeate flows and over-sizing may require excess
pressure to be burnt-off at the throttling valve.
An alternate option would be equipping the HP pumps with a VSD, while the
booster pumping system would be equipped with a fixed speed drive, only
providing sufficient pressure to overcome friction and head-losses within the
cartridge system, as well as ensuring adequate NPSH is available. Reasoning
behind not further pursuing this alternate option was informed by the following
matters:
The cost of large high voltage variable speed drives for high wattage HP
pumps is significant
Inclusion of a variable speed drive would (permanently) reduce the energy
efficiency of the associated drive by 2% to 3%
Page 50

Interim Report

Preference to have majority of pumping capacity driven by a fixed speed


drive and operating at the best efficiency point (BEP)
7.3

Cartridge Filtration
To protect the RO membranes, cartridge filtration will be provided for each RO train
downstream of the filtered water transfer pumps. The cartridge filters shall be of 5
microns.

7.4

Oxidant Control
The provision of sodium bisulphite (SBS) dosing is included in the design to
neutralize residual oxidants present in the feedwater due to intake shock dosing.
This will be controlled by measurement of oxidation reduction potential of the
SWRO feedwater. The details of the bisulphite dosing system are sated in
approaching sections.

7.5

Scale Control
Provision for scale control by means of antiscalant dosing is included in the design.
The sulphuric acid addition as part of the pre-treatment system will also assist with
prevention of scaling. Refer to chapter 9 for details of the antiscalant and acid
dosing systems.

7.6

Reverse Osmosis Pumping and Energy Recovery System


The energy recovery devices will be an isobaric design.

Either the Energy

Recovery International (ERI) or Calder DWEER technology could be used. The


design is based on the Calder DWEER technology but both suppliers are
considered acceptable.
Details of the booster pumps are shown in Table 15.

Page 51

Interim Report

Table 15 : RO and ERD Feed Booster Pumps Design Parameters


Parameter

Unit

RO Feed Booster ERD Feed Booster


Pump

Pump

No.

Total number of pumps for 150 No.

L/s

295

346

ML/d

25.5

29.9

bar

14.5

6.0

Variable speed

Variable Speed

Number of pumps per RO train

MLD module
Maximum flow capacity
(per train)

Delivery Pressure
(are dependent on final membrane and ERD
selection and performance)

Type of drive

Materials (PREN > 43 for contact with seawater or brine):


Casing

Stainless
(Super duplex)

Pump impeller

Shaft

Stainless

steel Stainless

steel

(Super duplex)
steel Stainless

steel

(Super duplex)

(Super duplex)

Stainless steel

Stainless steel

Design parameters for the high pressure RO pumps are shown in Table above.
Since the High Pressure RO pump is a fixed speed pump it is important that the
pressure capability is not too high as that could lead to significant power wastage
with new membranes and warm/low salinity feed water.

Page 52

Interim Report

Table 16

High Pressure Pumps Design Parameters

Parameter

Unit

Value

Number of pumps per RO train

No.

Total number of pumps

No.

Maximum flow capacity

ML/d

25.5

L/s

295

bar

55.3

Delivery Pressure*
Type of drive

Fixed speed

Materials (PREN > 43 for contact with


seawater or brine):
Casing

Stainless

steel

(Super

steel

(Super

duplex)
Pump impeller

Stainless
duplex)

Shaft

Stainless steel

Notes: * Pressure requirements are dependent on final membrane selection and


performance.

Design parameters of the proposed ERD system are shown in Table 17


Table 17

Energy Recovery System Design Parameters

Page 53

Interim Report

Parameter

Unit

Value

Number of ERD systems per RO train

No.

Total number of ERD systems

No.

Number of ERD units per system

No.

Total number of units

No.

40

Type of system

Isobaric System

Type of energy recovery device

DWEER

Manufacturer of device

Calder AG

Design flow rate (Brine)

MLD

6.0 (per DWEER unit)

L/s

69

Maximum available head for recovery

bar

62

Efficiency at design point

~ 95+

Materials (PREN > 43 for contact with seawater or brine):


Casing

Stainless

steel

(Super

duplex)
Piston
Other

parts

in

contact

with

Non Metallic
Stainless

seawater/concentrate

steel

(Super

duplex)

Table 18 ERD Booster Pumps Design Parameters


Parameter

Unit

Value

Number of pumps per RO train

No.

Page 54

Interim Report

Total number of pumps

No.

Maximum flow capacity

MLD

29.9

L/s

346

bar

5.7

Maximum pumping head


Type of drive

Variable speed

Materials (PREN > 43 for contact with seawater or brine):


Casing

Stainless

steel

(Super

steel

(Super

duplex)
Pump impeller

Stainless
duplex)

Shaft

7.7

Stainless steel

RO Rack Design
The membrane design parameters are shown in Table 19. Normal production will
be using 6 racks for 150 MLD module at a flux of approximately 13.5 LMH.
Table 19 RO Membrane Design Parameters
Parameter

Unit

Value

Number of RO membrane trains -Total

No.

Number of racks per train

No.

Page 55

Interim Report

Number of RO racks

No.

RO Plant Recovery

46

Material of membranes

Polyamide

Type of membrane/module

8 inch spiral wound

Membrane manufacturer

Hydranautics, Dow

Active area per membrane element

m2

41 (440 ft2)

RO membrane flux rate

Normal

LMH

13.5

Maximum

LMH

15.5

Salt rejection

Nominal

> 99.8

Boron retention

Nominal

> 93

Salt Passage Increase

%/a

10

Flux Decline

%/a

3.5

Number of pressure vessels per rack

No.

240

Number of membranes per vessel

No.

Total number of membranes per train

No.

1920

Total number of pressure vessels for 150 MLD No.

1440

Average

Membrane

Life

(AMLT)

permeate
Total number of membranes for 150 MLD No.

11520

permeate
Reserve space provided for installation of %

additional membrane modules


Feed Water SDI requirements

100th% ile <3.5

To allow trialling and testing of individual elements to determine performance, a


single one-element vessel will be installed and operated in parallel to one of the
RO racks.

Page 56

Interim Report

The piping material for the RO system is as shown in

Table 20

Table 20 Piping materials


Piping

Material

High pressure feed piping

Stainless

steel,

super

duplex, PREN > 43


High pressure feed and concentrate piping and Stainless
headers

super

duplex, PREN > 43

Connecting pipework feed/concentrate headers to Stainless

7.8

steel,

steel,

pressure vessels

duplex, PREN > 43

Product headers

Stainless steel

Product piping

Stainless steel

Piping for cleaning

GRP

Piping for shut-down flushing

GRP

super

RO Clean In Place System


A RO clean in place (CIP) system is included in the design. This includes a
chemical cleaning preparation tank, chemical cleaning buffering tank, a heat
exchanger, pumps, cartridge filters, cleaning chemical storage tanks, cleaning
network and cleaning recirculation loop piping and a neutralisation system.
Provision has been made in the design for storage and dosing of the following
cleaning and preservation chemicals:

Caustic soda

Sodium Bisulphite

Hydrochloric Acid

Phosphoric Acid

Citric acid

Detergents (surfactant)

Ammonia solution

Cleaning occurs intermittently and all of these chemicals would not be used at the
Page 57

Interim Report

same time. The amount and type of cleaning chemical required would vary
depending upon the degree of membrane fouling and the nature of the fouling.
Depending on the membrane selected, the system may be designed to enable
cleaning of the RO system from both directions, i.e. from the front to the back and
from the back to the front. This would allow employing specific cleaning
procedures for a particular type of fouling / scaling. The CIP system will be sized
to clean each individual RO rack separately. The system, including the chemicals
for the CIP will be located in a dedicated building located next to the main RO
building.
All wastes from the CIP process will be neutralized prior to being directed to the
outfall and for the same a Neutralization Tank has been provided.
After chemical cleaning and prior shutdown of membrane trains, the brine and
spent cleaning solution will be flushed out of the RO and ERD racks using RO
permeate. Flushing prior train shutdown will reduce potential for scale build up
and corrosion, and will also reduce fouling and the amount of high TDS water
produced on start up. The flushing system would consist of flushing pumps and
pipework allowing each RO train to be flushed individually.

Design parameters of the chemical cleaning preparation tank are shown in Table
21.
Table 21 Chemical Cleaning Preparation Tank Design Parameters
Parameter

Unit

Value

Number of tanks

No.

Net volume of tank

300

Body / ends

GRP

Pipes / valves

GRP/PVC/SS

Material:

Form of construction

Circular

Design parameters for the chemical cleaning pumps are shown in Table 22.
Table 22 Chemical Cleaning Pumps Design Parameters
Page 58

Interim Report

Parameter

Unit

Value

Number of pumps

No.

2W+1S

Flow capacity of each pump

L/s

300

Pumping head

bar

10

Number of stages

No.

Design parameters for the flushing pumps are shown in Table 23.
Table 23 Flushing Pumps Design Parameters

7.9

Parameter

Unit

Value

Number of pumps

No.

1W+1S

Flow capacity of each pump

L/s

295

Pumping head

Number of stages

No.

Permeate Storage
One permeate storage tanks will be located immediately downstream of the RO
plant. The tanks have been sized for at least 20 min storage at the maximum
permeate flow of 153 MLD. The design parameters for the Permeate Storage Tank
are summarised in Table 24.
Table 24 Permeate Storage Design Parameters
Parameter

Units

Value

Number of tanks

No.

Nominal volume per tank

ML/tank

2.125
Page 59

Interim Report

Page 60

Interim Report

8.

POST TREATMENT
Post treatment of permeate is required to meet the statutory product water quality
requirements.

Post treatment will consist of remineralisation/stabilisation and

disinfection of the water.


8.1

Re-mineralization/Stabilization

8.1.1

Requirement for Stabilisation


Water produced by a reverse osmosis process has very low residual hardness and
alkalinity, which renders it very aggressive to most materials including steel and
concrete thus causing corrosion and premature aging of assets. The lack of
carbonate alkalinity as well as the low content of calcium and magnesium (i.e.,
very low hardness) causes desalinated water to be unstable and prone to
wide variations in pH due to its low buffering capacity and its inability to form
protective calcium carbonate films on pipe walls, which makes it corrosive.
Therefore, before the permeate from reverse osmosis be supplied to customers, it
needs stabilisation also known as remineralisation or conditioning to prevent this
from happening.
Product water remineralisation for corrosion protection has two aspects:
(1) protection of water distribution systems and household piping and fixtures from
damage caused by corrosion, and
(2) maintaining the desalinated water quality in terms of aesthetic appearance
(color and taste).
Table 25

Stabilisation General Design Parameters

Item

Requirement

Stabilization Process

Limewater/Carbon Dioxide Dosing

Target pH range

7.5 to 8.5

Target Langelier Index

Positive

Maximum Total Dissolved Solids

500 mg/L

Page 61

Interim Report

8.1.2

Hydrated Lime & Carbon Dioxide


Water shall be stabilised by the addition of carbon dioxide and lime (calcium
hydroxide). The calcium is dosed via limewater, which is produced by mixing
powdered hydrated lime with reverse osmosis permeate using a clarifier. As the
lime available in the country contains impurities a considerable volume of waste is
generated by the limewater clarifier. This waste is relatively inert and could be
readily dried onsite.

Left to Right: Lime Silo, Limewater Clarifier, Carbon Dioxide Storage Vessels

Carbon dioxide gas is added to the water. It reacts with the dosed lime to form
calcium bicarbonate, which buffers the water and increases the resistance to
changes in pH and thus reduces the corrosivity of the water. Carbon dioxide and
lime are currently in common use all parts of the world for treatment of drinking
water supplies for the said purpose. The Perth Seawater Desalination Plant (130
MLD) uses the said process for portabalization.
Equipment typically required for dosing of lime would consist of: lime storage silos,
mixing vats, limewater clarifier, lime-water storage tank and dosing pumps. Carbon
dioxide would be stored and dosed using a proprietary package plant consisting of
storage vessels, evaporators and dosing venturis.
The remineralization using lime and carbon dioxide follows the chemical reaction
presented below:
2CO2 + Ca (OH)2 Ca (HCO3)2

0.74 mg/L of hydrated lime (0.56 mg/L of quicklime) and 0.88 mg/L of carbon
dioxide would need to be added in order to increase desalinated water hardness
Page 62

Interim Report

and alkalinity by 1.0 mg/L (as CaCO3)

each.

Therefore, for a target

recommended dosage of alkalinity and hardness in the product water of 100 mg/L,
the water produced by the desalination system will need to be treated with 74 mg/L
of lime and 88 mg/L of carbon dioxide.
The lime product used for remineralisation is usually delivered and stored at the
desalination plant site in silos as either powdered hydrated lime or as
pebble-lime (CaO), which is than slaked to generate hydrated lime [Ca(OH)2].
This lime slurry is blended with the fresh water produced by the desalination
process and is thoroughly mixed in lime saturator tanks to create saturated
limewater, which is then injected into the unconditioned desalinated water.
Carbon dioxide is typically delivered to the desalination plant site in a liquefied
form and is stored under pressure in metal storage tanks.
The majority of desalination plants are designed to produce finished water of total
(calcium and magnesium) hardness of 80 to 120 mg/L as CaCO3. Since
desalinated water usually contains less than 2 mg/L of magnesium, over 90
precent of the total hardness of drinking water conditioned with lime is calcium
hardness.
8.1.3

Key design Criteria


Lime Silo and Slurry System
Lime is delivered to the desalination plant site as bulk powdered hydrated lime,
which is stored on site in cylindrical metal silos with a 60 cone bottom (lime bins).
These silos are often the tallest structures on the desalination plant site and
usually have a diameter between 2.0 and 6.0 m with a height of 5.0 to 15.0 m.
They are equipped with air vent filters and with bin vibrators to promote flow of lime
from the silo into a transition bin hopper. The feed of lime is controlled by a rotary
discharge (airlock) valve, which is motorized and closed or opened automatically
based on the level of lime in the hopper. Lime silos are typically welded/bolted
stainless-steel structures equipped with level sensors, a fill line with long-radius
elbows, and a truck hose loading fitting with dust cap.

Carbon Dioxide System

Page 63

Interim Report

Carbon dioxide is delivered to and stored at the desalination plant in compressed


liquid form [20 bars] of pressure at 20C. A typical carbon dioxide feed system
consists of liquid carbon dioxide storage tanks equipped with refrigeration units,
vaporizers, vapor heaters, carbon dioxide dosing system with pressurized gas
solubility panel, and an in-line gas feed diffuser.
Table 26

Carbon Dioxide Design Parameters

Component/Parameter
Specifications/Design Criteria
Carbon dioxide dosage (as 100% For lime/CO2 systems88 mg/L For
pure product)

calcite/CO2 systems44 mg/L per


1.0 mg/L of target alkalinity and

hardness concentrations (as CaCO3)


CO2 consumption (kg/day as 70.4105.6
100% pure product) per 1000 m
3 /day of desalinated water for
addition

of

alkalinity

and

hardness in a range of 80 to 120


m/L (as CaCO3
CO2 purity, %
Liquid CO2 storage

100
tank

(refrigerated)
Storage pressure/ temperature, 20 bars/20C
bars/C
Diameter, m
Length/Height, m
Storage time, days
Structurematerial type
Vaporizer
Type
Capacity,
horsepower/kg

0.83.0
515
1560
Carbon steel
Electric
0.030.05

CO2.day
Delivery pressure/temperature, 3 bars/12C
bars/C
Vapor Heater
Type
Capacity,

Electric
horsepower/kg 0.010.02

CO2.day
Key design Criteria Carbon Dioxide and Feed System
The dosage of carbon dioxide fed to the desalinated water is adjusted
Page 64

Interim Report

proportionally to the desalinated water flow rate by flow control valves installed on
the gas feed lines from the vaporizer to the point of chlorine dioxide gas injection in
the desalinated water conveyance pipe.
8.2

Disinfection

Disinfection has been used for over a hundred years, when it was first regularly
used to kill germs in the 1900s. Chlorination was first used in London in 1905 and
ozone was first used in Nice in 1908. From there, disinfection developed further
and was widely accepted as a method to treat water. Chlorination is the most
widely used method with ozone slowly becoming more popular. Chlorine dioxide is
also another disinfectant in the water treatment industry, but because of its high
cost not used commonly. UV disinfection is a relatively new disinfection method
even though it was also discovered approximately at the same time as the other
techniques.
Requirement for Disinfection
Disinfection is the process whereby water is treated to kill microorganisms in the
water, including essentially all pathogenic (disease-causing) bacteria. In addition
disinfection is undertaken to prevent bio-film (slime) growth inside the pipes and
mitigate the risk of recontamination.
Chlorine based disinfection (i.e. chlorination) has been considered for said project.
This is because this process is by far the most common form of disinfection in the
country and has a shell life as compared to other disinfection process which
usually doesnt have the required shell life in distribution system.
Chlorine is the most widely used water disinfection method. Chlorine kills the
micro-organisms by immobilising their metabolism rendering them harmless.
Chlorine is a slow stable reaction thus its main advantage of chlorine is the
formation of residuals which remain in the water for longer periods of time
protecting the system from bacterial contamination.

Necessity for Disinfection of RO Process Water


The need for continuous disinfection of the process water has been questioned,
since reverse osmosis permeate will have a relatively low/nil level of
Page 65

Interim Report

microbiological contamination. The reverse osmosis treatment process will remove


the majority of organic matter present. This results in a limited amount of short
chain carbon compounds available as a food source for micro-organism growth.
However to have a residual chlorine in the distribution network chlorination is
envisaged for the treated water out of the proposed plant.
Theory of disinfection and Chlorination
The rate of change in the number of viable organisms is the product of the reaction
rate constant for particular disinfectants and the number of viable organisms
(Tebbutt, 1992). It is expressed as:
dN/dT = - kN

Where k = rate constant for particular disinfectants


N = the number of viable organisms.
This expression is also defined as Chicks Law and is the first-order reaction of a
uni-molecular reaction (Crittenden et al., 2005). Integrating the formula and
changing to base 10 gives the formula:
Log (Nt/No) = - kT
Where;
Nt =

Number of micro-organisms at time t

No =

Number of micro-organisms initially

0.4343 K

Page 66

Interim Report

Rearranging the formula to determine the time required for the adequate disinfection is
expressed as:
t = (1/k) log(No/ Nt)
Since the number of micro-organisms can never be zero no matter how much
disinfected the water is, the percentage of removing micro-organisms can never be 100%.
However, majority of the water treatment plants must be able to remove or inactivate
99.9% (3-log) of Giardia cysts and 99.99% (4-log) of viruses (Critttenden et al., 2005).
Some micro-organisms are more vulnerable and have less resistance to certain
disinfectants therefore only 90% (1-log) reduction is required.
For effective disinfection there are numerous parameters that are required for killing
or inactivating micro-organisms. This includes the residual concentration, the time at
which the water is in contact with the disinfectant, the water temperature, the pH balance,
and the type of disinfection used.
If the concentration of disinfectant, C, measured in milligrams per litre (mg/l), and
the contact time, T, measured in minutes, are multiplied together it will produce a
value known as the CT value. This value is required to produce adequate parameters in
order to achieve the required log requirements. Different micro-organisms will require
different CT values therefore, in general, the contact time and the concentration of the
disinfectant must be able to remove the most resistant of the micro-organisms. It is
important for the water system to have an adequate CT value so that the water is treated
properly before being distributed.
In alkaline conditions, chlorine is more
effective and in acid conditions, the
hypochlorous acid is more effective.
Therefore it is recommended that the
typical pH for disinfection should be
lessmthan 8 and if chlorine is used,
then a lower pH is better.
Chloramine, which can be formed
during

chlorine

and

ammonia

Interim Report

chemical reaction, is a combined residual chlorine and is also an useful disinfectant.


Break point occurs once all the ammonia is reacted and the combined residual
chlorine is converted back to free chlorine (Refer figure above). Increasing the chlorine
concentration also increases the free-chlorine residual, however increasing it beyond the
break point is counterproductive due to the production of nitrogen and nitrogen trichloride, chemicals that have a strong unpleasant odour.
The break point is a good indicator when to supply sufficient amounts of chlorine to remove
any excess ammonia without the risk of producing less welcoming products including dichloramine and nitrogen trichloride.

Chlorine Dosing Considerations


It is important that personnel, engaged in chlorine plant or in any activity involving handling
of chlorine, should understand the hazardous properties of chlorine and should have a
good grasp of the basis of preventive measures. This will make them more conscious
about safety which would not be achieved if they are trained to follow certain instructions
mechanically. In India, Chlorine is deemed to be an explosive, when contained in any
metal container in a compressed or liquefied state, within the meaning of the Indian
Explosives Act, 1884. The filling, possession, transport and importation is governed by the
Gas Cylinder Rules, 1940.
Cylinders should be stored in an upright position. They should be
secured to prevent from falling over.
Full and empty cylinders should not
be stored together. Then containers
should be stored on their .sides.
They should not be stacked or
racked more than one high. All
safety measures shall be as per IS
4263 latest edition Code of Safety
for Chlorine and IS 10553 (Part I)
1983 Requirements of Chlorination
equipment for its safe operations. Also according to the Compressed Gas Cylinder Act

Interim Report

Government of India, it is necessary to obtain a storage license when five or more


containers are to be stored. For obtaining the storage license safety equipments are
essential along with scrubber system. Also the building has to be constructed as per
norms suitable for storing toxic gas cylinders with adequate ventilation systems and
handling accessories. Further as per Manufacture, Storage and Import of Hazardous
Chemicals Rules, 1989 maximum of 10 tonnes can be stored at site
Chlorine Dosing Rates
The recommended chlorine dosing rates for Potable water are;
Continuous disinfection of potable water using gaseous chlorine at a concentration
ranging between 2.0 3.0 mg/L, to achieve a residual chlorine of 0.2 mg/l as per IS
10500.
Disinfection of the produced water will be carried out using a standard chlorine gas
system. Both the process and potable quality product waters will be dosed. To
achieve is minimum Residual chlorine in distribution system, 2 mg/l of chlorine shall
be dosed using 20 kg/hr. (1 W + 1 S).
Table 27: Liquid Chlorine Characteristics
Properties
Appearance

Greenish Yellow

Physical state

Liquefied compressed Gas

Vapour pressure @ 35oC

27.48 mm Hg at 20oC

Boiling point /Range

-34 C (at 1 Atmospheric Pressure)

Melting point/Freezing Point

-101C

Vapour Density (Air Density= 1)

2.49

Liquid Density(Water density = 1)

1.47 at 20oC

Solubility in water

0.7 % at 20oC

Odour

Irritating, Bleach like Choking Odour

Others

Soluble in alkalis

Specifications
Chlorine Content Min.V/V

99.8%

Moisture

150 ppm Max.

Residue on Evaporation

Shall leave no residue when


evaporated on clean white tile

Interim Report

Specification of Chlorine Tonner


Water Capacity (approx.) Kg.

780

Chlorine Capacity (approx.) Kg.

930

Design Pressure, Kg/cm

9.

19.9

Inside Diameter (approx.) mm

760

Shell Thickness, mm

10

Dished Ends Thickness, mm

9.6 (Min.)

Overall Length (approx.) mm

2085

Tare Weight (approx.) Kg.

620

Valve Outlet

5/8

Thread

14 tpi bsp

CHEMICAL SYSTEMS
The following chemicals will be required as part of the pre- treatment process:
Table 28 Desalination Plant Chemicals List
Chemical Name

Chemical Use

Antiscalant

RO scale prevention

Carbon Dioxide

Post treatment for water stabilisation

Interim Report

Chlorine Gas

Post

treatment

for

disinfection

(process & potable product waters)


Ferric Chloride

Coagulant

Hydrated Lime

Post treatment for water stabilisation

Flotation Aid - Polyelectrolyte

DAF flocculation aid

Limewater clarification aid

RO CIP

Clarification/Filtration

Aid

Polyelectrolyte
Sodium Hydroxide
(Caustic Soda)
Sodium Hypochlorite
(prepared

Preparation of sodium hypochlorite

solution
Shock dosing for intake biofilm control

Protection of RO from oxidants

from

Sodium chlorine/hydroxide)
Metabisulphite

Sulphuric Acid

RO CIP
pH adjustment and scale control

RO CIP only
Miscellaneous Chemicals
(eg: Biocide, Detergent, Ammonia, Acids: Hydrochloric, Phosphoric, etc)

Details of the chemical systems are summarised in the following sections

9.1

Antiscalant
Antiscalant is required to be dosed into the reverse osmosis system to manage the
formation of mineral scale on the surface of the membranes. Design parameters of
the antiscalant system are shown in Table below.
Table 29 Antiscalant System Design Parameters
Parameter

Units

Value

Design RO system feed

ML/d

356

Interim Report

flow rate
Antiscalant concentration

Form of supply

100
1000 L IBC

Dose rate

mg/L

Dosing point

0.7
RO

High

Pressure

Pump suction main


Antiscalant consumption

kg/day

Specific gravity of dosing

250
1.2

solution
Dosing

flow

rate

of

L/hr

Duty

No.

Standby

No.

Dosing pump capacity

L/hr

Number of antiscalant storage tanks

No.

None

antiscalant
Number of dosing pumps

direct

feed

from IBC

It is not proposed to dilute the antiscalant as the flow rates involved can be easily
dosed accurately and dilution can lead to degradation of the antiscalant. Adequate
mixing will be achieved by other means. At this stage the type of antiscalant to be
used has not been determined. This will be assessed at the next design stage.
9.2

Carbon Dioxide
Carbon Dioxide is required to be dosed in association with Hydrated Lime for pH
control and stabilisation of the permeate water. The chemical dosing design
parameters are summarised in Table 30.
Table 30 Carbon Dioxide Dosing Design Parameters
Carbon Dioxide Dosing

Units

Value

Interim Report

Carbon Dioxide Dose Rate


Maximum

mg/L

90

Average

mg/L

70.4

Carbon Dioxide Dose

Average

kg/hr

575

Carbon Dioxide Vessels

Number Of

No.

Capacity

tonnes

25

days

Carbon

Dioxide

Storage

Period
9.3

Chlorine Gas
Chlorine is required for disinfection of the product water and for shock chlorination
Chlorine will be supplied as liquefied gas in 900 kg drums. The unloading of
chlorine drums from transport trucks will occur within the chlorine building, thereby
reducing the risk of a chlorine leak occurring outside the building. The system will
be configured as a vacuum gas draw system with vacuum valves connected
directly to the chlorine drums.

Design parameters for the proposed chlorination system are summarised in 31.
Table 31 Chlorination Design Parameters
Parameter

Units

Value

Minimum

mg/L

Maximum

mg/L

Treated Water Flow Rate

Nominal

ML/d

153

Chlorine gas flow range

Nominal

kg/hr

12.75

Maximum

kg/hr

19.125

Maximum

kg/day

459

Average

kg/day

306

Chlorine for Disinfection of Potable Water


Dose Rate

Chlorine Consumption

Chlorine Storage

Interim Report

Total Chlorine Consumption

Maximum

kg/day

459

Chlorine Container Size

kg

900

Maximum Number of Drums Online

No.

days

15

On Site
Drum Storage Period

9.4

average

Ferric Chloride
Ferric Chloride is to be dosed to coagulate particles prior to the DAF. The design
parameters are summarised in Table 32.
Table 32 Ferric Chloride System Design Parameters
Parameter

Units

Value

Design seawater flow

ML/d

395

Ferric chloride concentration

% w/w

42.6

Specific

gravity

of

dosing

1.45

Solution
Dose rate

Ferric Chloride Consumption


Dosing Points

9.5

Typical

mg/L

16

Maximum

Mg/l

25

Typical

t/day

15.0
Flash Mixers U/S of
DAF

Hydrated Lime
Lime will be supplied to site as hydrated lime (Ca(OH)2) and converted onsite to

Interim Report

lime water before dosing to permeate upstream of product water. Lime does not
dissolve quickly in water and is difficult to handle. Lime is mixed with water to
obtain lime water, and then the lime water is passed through a clarifier to remove
the un-dissolved lime; if not removed the product water will not meet specification.
Equipment for handling lime requires regular maintenance, and adequate standby
provisions must be made. Two lime water clarifiers will be provided. Lime is not
pure and significant amounts of sludge are produced and must be dewatered and
removed from site. The commercial lime available has a purity of 80% or less; a
purity of 75% has been adopted for design.
The lime dosing design parameters are summarised in Table 33.

Table 33

Lime Dosing Design Parameters

Parameters
Lime Properties

Units

Value

75

Utilisation of pure %

90

Purity

component

Lime Dose Rate (Total)

Lime Consumption

Lime Silos

Utilisation overall

67.5%

Typical

mg/L

50 as Ca(OH)2

Maximum

mg/L

75 as Ca(OH)2

Average

tonnes/day

11.35 t/day

Maximum

tonnes/day

17 t/day

Number

No.

Capacity

of

each tonnes

50

silo
Lime

of

Milk

Concentration
9.6

Flotation Aid Polyelectrolyte


Polyelectrolyte is required to be dosed to assist with the formation of floatable flocs
in the DAF process. The polymer will be dosed immediately prior to each DAF

Interim Report

flocculation tank. The specific polymer to be used will be determined during a later
design stage. The dosing design parameters are summarised in table 34

Table 34

Flotation Aid Polyelectrolyte Dosing Design Parameters

Parameters

Units

Value

Design seawater flow

ML/d

395

% w/v

0.1 0.5

Poly Solution Strength

Nominal

Dosing Location

U/s of Lamella
U/s of DAF

Dose Rate (per dosing Minimum

mg/L

0.05

Maximum

mg/L

0.25

Average

mg/L

0.1

point)

9.7

Sodium Hydroxide
Sodium hydroxide is required to neutralize acids used in the cleaning of the UF
membranes and may also be used for cleaning of the RO membranes. Sodium
hydroxide may also be required to increase the pH of the RO feed water in order to
meet boron target levels.

Table 35 Sodium Hydroxide Dosing


Parameter

Units

Value

Hydroxide concentration

% w/w

50

Specific gravity of dosing solution


Storage Details

1.53

Interim Report

9.8

No. of Tanks

No

Tank Volume

KL/tank

150

Dosage location

Inlet of Cartridge filter

Dosage

25 m/l average

Sodium (Meta)bisulphite
Sodium bisulphite (SBS) (NaHSO3) will be prepared by dissolving sodium
metabisulphite (SMBS) (Na2S2O5) in water. Dosing of bisulphite is required to
remove residual chlorine from the system resulting from intake shock dosing as
well as to neutralise oxidants used for cleaning of the UF membranes. Bisulphite
may also be used for the RO CIP process.
The redox potential (ORP) will be measured downstream of the dosing point to
ensure that the dosing is effective. The sodium bisulphite design parameters are
summarised in Table 36.
Ventilation shall be provided to manage any fumes that may be evolved in the
chemical area.
Table 36 : Sodium (Meta)bisulphite System Design Parameters
Parameter

Units

Value

Sodium bisulphite concentration

% w/v

40

Specific gravity of dosing solution

1.34

Dosing for RO Protection


Design seawater flow
Dose rate (as sodium metabisulphite)

Dosing frequency / duration

ML/d

395

Maximum

mg/L

30

Typical

mg/L

15

Assume 1 hours once per day (at


time of intake shock)

Dosing point

Upstream of high pressure pumps

Interim Report

9.9

No of Dosing pump

9 W+4 S

Capacity of each pump

125 L/hr

Sulphuric Acid
The sulphuric acid will be used to provide pH correction for optimum seawater
coagulation. In addition sulphuric acid is required for chemical enhanced
backwashing and cleaning of the UF membranes. The design parameters are
summarised in Table 37.
Table 37

Sulphuric Acid System Design Parameters

Parameter

Value

Sulphuric acid concentration


Specific

gravity

of

% w/w

dosing

98
1.84

solution
Seawater pH Correction
Design seawater flow

395

Dose rate

Typical

35

Dosing flow rate of chemical

Typical

350 L/hr.

Dosing point

Seawater

intake

pipeline

upstream of coagulant dosing

Interim Report

9.10

Ultra Filtration Chemicals

Table 38

Ultra Filtration chemical Usages

Parameter
Oxidant

Value
backwash

with

Once a day

NaOCl (10%)
Number of tanks

One

Volume
Acid

M3 300

backwash

with

HCL

Once every 7 days

(35%)
Number of tanks
Volume

9.11

One
m3

Proposed Acid

HCl

Proposed Alkali

NaOH

Proposed Oxidant

NaOCl

Dosing Frequency

Once in 60 days

CIP tank Volume

5 m3

Miscellaneous Chemicals
This section covers miscellaneous chemicals used only for the purpose of cleaning
the RO membranes (i.e. RO CIP). Examples of these chemicals include (but may
not be limited to):

Biocide;

Detergent;

Ammonia solution; and

Acids (such as Hydrochloric, Phosphoric, etc)

Interim Report

General storage space only has been included for these chemicals because:

It will not be known until the first year (or two) of operation which of the
chemicals are required;

The chemicals are used infrequently (say once per year); and

When used, only small volume of chemical is used each time.

The facility will be designed to meet the requirements of all of the chemicals that
may be used. Between uses, the facility will need to be drained and flushed.
Design details are provided in the following table.
Table 39 Miscellaneous CIP Chemical Dosing System Design Parameters
Parameter

Units

Value

Dosing point

RO CIP Tank

Specific gravity of dosing solution

varies

Number of dosing pumps

Duty

No.

Standby

No.

Dosing pump capacity

kL/hr

Number of tanks

No.

Tank volume

kL/tank

Chemical delivery volumes

9.11.1

Multiple 1000L IBCs

Cleaning-in-place (CIP)
To manage membrane fouling and scaling, periodical cleaning is required.
Depending on the performance of the pre-treatment system and on the actual
operational envelope, it is anticipated that four cleanings per year per train are
necessary. Typical cleaning solutions include low pH, and high pH solutions and
special detergents. Preservation solution would also be prepared within the CIP
system.

To facilitate membrane cleaning a CIP system is required with the following

Interim Report

features:

Make-up tank;
Chemical storage and dosing facilities for the preparation of various cleaning
solutions (typically includes Caustic Soda, Citric Acid, NaDDS, Sodium
Bisulphite);

Recycling pumps;

Cartridge filter (typically 5 micron);

Pipework allowing each RO train to be cleaned individually; and

If required, holding tank to neutralise spent cleaning solution and enable


discharge to environment.

9.11.2

Flushing
After chemical cleaning and prior shutdown of membrane trains the brine and
spent cleaning solution should be flushed out of the RO and ERD racks using RO
permeate. Flushing prior train shutdown will reduce potential for scale build up and
corrosion.
The flushing system would consist of flushing pumps and pipework allowing each
RO train to be flushed individually.

Interim Report

10. WATER STORAGE AND TRANSFER


10.1

Process water storage


Treated process water will be stored on site prior to being transferred to the city for
consumption. The criteria for the storage tank are summarised below;

10.2

Total usable storage volume 6250 M3

Number of units : 1 x 6.25ML (1 hours storage for 150 MLD average flow),

Material of construction : RCC epoxy coating

Tank geometry : to suit site hydraulics and cost optimisation,

Covered

Process water transfer pump station


The design criteria adopted for design of the pump station is therefore as follows;

Type of pumps : centrifugal variable speed drive,

Number of units : 3 duty, 1 standby (for 150 MLD),

Nominal pump station duty : 2085 m3/hr,

Pump control : by level in the receiving tank,

Positive suction at all times,

11.

SEAWATER OUTFALL

11.1

Outfall Pipeline
The desalination plant will discharge rejected seawater concentrate and other
waste streams into the ocean via an outlet pipeline.
The criteria adopted for Section 1 of the outlet pipeline are as follows;

Interim Report

Peak Flow : 242 MLD,

Diameter: 1400 mm ID RCC pipe

Max working pressure (including water hammer allowance):


20m head,

Note *
The head required to discharge seawater concentrate is provided by the residual seawater
concentrate pressure after the reverse osmosis process. This pressure would otherwise be fully
recovered and therefore has to be supplemented to make up for the loss of head to the outlet
pipeline. The outfall pipeline can therefore be considered as consuming energy (as with a pump
station) with an associated operating cost.

11.2

Diffusers
The outlet pipeline will discharge seawater concentrate and other waste flows into
the ocean via a system of diffusers. These comprise a large number of small
diameter nozzles discharging jets of seawater concentrate from the seabed into
the water body to entrain dilution flows. Wider dispersion of the saline plume is
achieved through the action of wind, waves and currents. Both the seawater
concentrate jets and the dispersion will be modelled using appropriate computer
models. The adopted criteria for the diffuser design is summarised as follows;

Flow range @ peak : 242 MLD,

Discharge temperature range: 0C > 3C above ambient,

Discharge

salinity

range:

32,500mg/l

-34,500mg/l

above

ambient,

Target dilution: 5% above ambient salinity,

Water depth: > 5m with no surface expression of seawater


concentrate jet,

Diffuser material: rubber / EPDM,

Backflow prevention: In line rubber check valve at diffuser,

12. POWER AND CONTROL


12.1

Power supply

Interim Report

12.1.1

Background
The power supply arrangement for the desalination plant is based on provision of
one nos (1 no.) 110 kV feeders from a substation next (very adjacent) to the
existing Neemlai SWRO to be procured and operated by TNEB.
The major areas for power demand are the seawater supply pumps, the RO
treatment plant and the process water transfer pump station.
The substation for the proposed plant has been placed on the north end of the site
along the ECR front, since the area embarked for the plant doesnt have sufficient
space to accommodate the electrical sub-station.

12.1.2

Design Considerations
It is anticipated that the concept design and cost estimate of the HV infrastructure
and electrical plant will include and take into consideration the following:

System security;

System reliability;

Determination of appropriate voltages and load currents for the main pump
motors and other large motors as appropriate;

Redundancy requirements;

Electrical supplier requirements for connection and design of substations;

National Electricity Law;

National Electricity Rules;

Energy losses;

Environmental impacts;

Lifecycle costs.

A level of security and redundancy will be determined for the reference design.
The appropriate Indian Standards will form the basis for the design of electrical
works including standards for:

Interim Report

Earthing

Protection

Switchgear, etc.

IEC Rules requirements will also be considered in the design.


12.1.3

110 kV / 11 kV Substation within the Plant premises


A 110 kV / 11 kV transformer shall be provided for the electrical supply to the
seawater intake pumps, the reverse osmosis high pressure pumps and the
process water transfer pumps. The L.T supply shall be from HT Panels. This
substation will be a double bus bar Gas Insulated System type (GIS) built with the
latest technologies, maximum operational safety and environmental friendliness
according to IEC 61850 and TNEB. Two incoming lines, coupler and four feeders
will be considered in the substation as minimum.
Two transformers 110/11 kV, ONAN / ONAF, will be installed and it is necessary to
consider civil works space for other two equal transformer (similar than
considered). The capacity of one transformer will be enough to feed the total
demand. Besides, it will contain all common equipment, no-load compensation and
necessary maneuvers considering three feeders in 11kV from spare transformer as
minimum.

12.1.4

Pumping station electrical works, controls and electrical systems


Voltages and load currents for the RO high pressure pumps and process water
pump motors (and any other large motors) will be 11 kV. Soft start or VSDs will be
provided for large motors (>400 kW).
11 kV Cables will be XLPE insulated, PVC sheathed, screened and comply with
the IEC 502.1. Low voltage power cables will be XLPE insulated and PVC
sheathed:
CU/XLPE/PVC type.
CU/XLPE/PVC type with concentric conductor, specific for frequency
converter outputs.
Rated insulation voltage: 1000 V.
The cores will be of copper.

Interim Report

Minimum cross section: power 2,5 mm


Power factor measured in 11kV will be more than 0.95 with filters if necessary.
12.1.5

Emergency power generation


Battery backed UPS systems will be provided for instrumentation, telemetry,
emergency lighting and communication systems.
The DC and UPS system shall be redundant (2 x 100% batteries, 2 x 100%
220VDC rectifiers, 2 x 100% 24VDC converters, 2 x 100% 230A inverters), DC
system shall be sized for a minimum period of 2 hour of standby operation.
One emergency diesel generator (EDG) set shall be provided. The EDG shall be
mounted in an outdoor soundproof metallic canopy with anti-vibration supports and
a high grade exhaust silencer. It will be designed for 24 hours continuous operation
and maximum 250 hours operation per year.

12.1.6

Electrical Equipment
Electrical equipment design will be in compliance with IEC Standards.
The electrical cable trays shall be GRP.
Cables of different voltages will generally be laid on separate trays. Where this is
not possible, cables may be installed on the same trays with suitable barriers.
The transmission cables of analogical signals will be laid on trays different than the
power cables to avoid any electromagnetic interference.
Where there is a possibility of mechanical damage or fire hazard, cable trays will be
protected by sheet steel covers. The cable inlet to the equipment will be protected
with tight stuffing boxes IP-65, which guarantees a perfect tightness.

12.2
12.2.1

Control System
General Design Criteria
The control system for the desalination plant will be compatible with the overall

Interim Report

control system and the design will take into account the following requirements:

The Process Control System (PCS) architecture covering the connection of field
instruments, controllers and associated input/output (I/O), Human Machine
Interfaces (HMI), communication networks and network equipment will be
designed for a process plant availability of at least 98%. This implies that that
the control system itself will have a system availability of at least 99.9%;

Use of equipment with a proven track record;

Provision of redundancy for critical items;

Redundant controllers, networks, power supplies in all locations;

Ensuring maintainability of selected equipment and standardisation of equipment


and spares;

Fibre optic will be provided as the primary bearer to link all units.

Process controls and equipment designed to fail to the safest condition;

I/O & instrument voltages selected to suit equipment and provided with battery
backup.

All on site power, control and communication cables to be run in accordance with
current industry practices, codes, standards and regulations;

Uninterruptible power supply (UPS) provided to all critical items (instruments,


control systems and telecommunications systems with a one hour battery backup with emergency power supply to UPSs within 15 minutes).

Field instrumentation will be rated IP56 and will have outdoor housing protection
from weather.

Lightning protection will be provided for all facilities. Surge protection is to be


considered for all facilities to provide protection to communications and other
electronic based equipment.

12.2.2

Plant Operation Methods


The following aspects of the plant operation will be taken into account in the design
of

the

control

communications.

system,

alarm

system,

supervisory

system

and

system

Interim Report

All plant operations will be designed for automatic operation and minimal
requirement for manual intervention;

The desalination plant and process water pump station will be monitored 24
hours per day;

Outside of the manned hours of the desalination plant and pump station,
operations or maintenance will only be in response to automated alarm
notification to nominated personnel on call;

For backup to the automated alarm notification outside of manned hours, the site
will also have 24 hour monitoring of all operating parameters and alarms;

Operating protocols will be established such that only the desalination plant will
have control ownership of the complete water supply system.

12.2.3

Control System Architecture


The following are the minimum requirements of the control system:

The desalination plant will be controlled by its own stand alone control system.
This will include all equipment, cabling, buildings and services necessary for a
complete system;

The Process Control System (PCS) will be of the same manufacturer, make,
model, firmware as that used for the magnetite process plant and will follow the
same configuration/ programming criteria and standards;

The Controller architecture, communication protocol and cabling selected will be


suitable for the application and control system requirements in terms of
functionality, speed, durability, off site monitoring and control requirements,
serviceability, and local support;

The HMI interface system on site will be suitable for the application, control
system requirements in terms of functionality, speed, durability, local support;

All HMI systems will display alarms, alarm histories and trends. All data must be
stored for more than twelve months;

All PCS and communication systems will have at least 20% spare memory, I/O
and communication capacity;

Each controllable drive will have a local control station comprising an emergency

Interim Report

stop (hard wired in the emergency stop circuitry for the drive which is not
disabled in auto mode), a local stop button and a local start button (for starting
the motor under maintenance /manual mode of operation). This local control
station will include a jog button when appropriate for some mechanical
equipment;

The PCS will incorporate monitoring of key alarms for other plant auxiliaries such
as the main low voltage electrical reticulation, building services and fire
suppression system.

Interim Report

13. CIVIL, BUILDING AND STRUCTURAL WORKS


13.1

Plant Buildings

13.1.1

General
The desalination plant will include the following buildings:

Intake Pump House

DAF (Dissolved air flotation) building

UF Building inc. UF fed Pumps and Self-Cleaning filters

RO Plant Building (inclusive of Cartridge Filters pressure vessels, booster and


HP pumps and energy recovery);

Administration and Control Building;

Chemical Dosing Buildings;

Chlorine Building;

Electrical Switchgear Buildings / Substation.

Buildings will generally be of concrete foundation slabs, sheet metal clad, insulated
and suitable for cyclonic conditions with a design life of 30 years. Buildings shall
comply with the relevant Indian Building Code.

Buildings will be provided with internal power, communications, lights, air


conditioning, ventilation, water, air and sewerage disposal facilities as well as
equipment and furnishings to provide fully functional facilities.
The Finished Floor Level (FFL) shall be a minimum of 300 mm above the external
finished ground level. External parking, laydown and walkway areas shall be
landscaped and illuminated.

Interim Report

Equipment buildings will be provided with appropriate doors to suit the operation of
the plant including removal and replacement of all equipment. Some buildings will
be provided with precast concrete walls to reduce noise levels and provide heat
insulation.
13.1.2

Platforms, Access Ways and Stairways


Platforms, access ways and stairways will be included in the design to allow
sufficient access for operations and maintenance.

Ladders will be avoided

wherever possible.
13.1.3

Lifting Devices
Lifting devices will be included in the design where necessary. These will include
electrically operated Cranes (EOTs) for removal of pumps and air blowers etc.
Gantry cranes will be provided RO buildings and the Chlorine building. Mobile
cranes will be required to move pumping and other equipment to a suitable truck
loading area.

13.2

Road works
The desalination plant area will be designed as a large, evenly graded area with
provision for adequate access to the plant such that vehicles will not need to cross
open drains. Grading around the plant will be arranged to ensure drainage away
from the plant at an appropriate grade to minimise the potential for ponding or
flooding.
Working areas around the desalination plant will be surfaced in a road base or
appropriate capping layer to provide all-weather hard standing areas for
maintenance traffic. Areas of significant traffic, such as around workshops and
offices and roads between facilities plus roads surrounding and internal to the
process plant shall be bitumen sealed. Any areas that are sensitive to vehicle
impact, in particular building edges, workshop door openings etc, exposed items
such as fire hydrants and other vulnerable infrastructure, shall be protected by 150
mm diameter steel bollards painted or marked (in reflective material) and, filled
with concrete or by roadside type guardrail.

Interim Report

All underground electrical cable in ducts /conduit will have concrete cable pits at
changes of direction. These pits will be provided with removable concrete covers at
plate level capable of taking the load from the outrigger of a 50 t rough terrain
crane. The covers will be levelled to the fished road level.
All Internal roads shall have a minimum carriage width of 6.0 m. Roads will be
wider in areas adjacent to chemical storage facilities to allow a chemical truck to be
unloading without restricting through traffic. Road widening will also be required at
bends to provide sufficient room for turning of a trailer. Road cross fall shall
generally be designed at 2% for sealed pavements and 5% for unsealed with
vertical and horizontal alignment to be designed in accordance with MORTH
publications.

13.3

Site Services
All electrical and communication services will be located in suitable underground.
Site pipework will be buried where feasible.

13.4

Storm water Drainage


Local drainage design around buildings will be based on a 1 in 5 year average
recurrence interval. Collector drains will be based on a 1 in 20 year return period
and over-all site drains will be designed for 1 in 100 year rainfall intensity. Drainage
flows will generally be handled by drains mostly covered. The drainage design will
be checked for a 1 in 100 year average recurrence interval rainfall event to ensure
that no facilities (including electrical equipment) suffer any permanent damage
from such a storm.
Storm water drainage from all facilities shall be contained and directed into run-off
channels. Drainage from process overflows will be returned back to the process.

13.5

Fire Systems
Fire services will be compatible with the requirements of Health, Safety and
Environment and relevant standards.

Interim Report

14. PRODUCT WATER MAIN


Sound planning dictates the standard design criteria in the planning and design of
pumping/ transmission mains. Pipelines are designed in accordance with the
maximum anticipated flow conditions; hence pipelines need to be sized to carry
water during these events without excessive pressure losses.
A water transmission system is the physical works that deliver water from the water
source to the intended end point or user. It is designed to deliver sufficient water
quantity and quality to meet the requirements of the customer. Typically, this is
achieved by way of pumps and motors, water mains, service pipes, storage tanks
or reservoirs, and related equipment, in a closed system under pressure. The
transmission

main

is

normally

designed for the carrying capacity


needed to supply water demand on
the maximum consumption day at a
constant rate. All hourly variations in
the water demand during the day of
maximum

consumption

are

then

assumed to be evened out by the


service reservoir.
A water system has two primary requirements: First, it needs to deliver adequate
amounts of water to meet consumer consumption requirements plus needed fire
flow requirements. Second, the water system needs to be reliable; the required
amount of water needs to be available 24 hours a day, 365 days a year.
In a pipe unlike the free-flow, the hydraulic gradient and not the slope of the conduit
will govern the flow. The hydraulic laws of closed conduit flows, also commonly
called pressurized flows, apply in the design of the same. Pressurized pipelines
can be laid up- and downhill as needed, as long as they remain at sufficient
distance below the hydraulic grade line, i.e. a certain minimum pressure is

Interim Report

maintained in the pipe. Hence, for closed conduits the hydraulic grade line slopes
according to the head loss.
Therefore, pipe sizing is based upon taking consideration of following factors:

Quantum of flow ( Q in m/s)

Fluid velocity (V in m/s)

Pressure (m in terms of water column)

Head loss (m in terms of water column)Pipe material

C-factor (Hazen-Williams coefficient of friction dependent on pipe material)

The governing equation is given by,


D= [4*Q/(*V)] 1/2
Where,
D= Internal Dia. Of pipe (m)
Q= flow (m/s)
V= velocity of fluid (m/s)
Transmission by pumping is applied in cases when the water has to be transported
over large distances and/or to higher elevations. The pumping head is the total
head, and comprises the static head plus the friction head loss for the design flow
rate. The pump to be selected must be able to provide this head at rated flow.
The head loss corresponding to the design flow rate can be computed for several
pipe diameters using Modified Hazen William Equation.
h = [L (Q / CR) 1.81]/ (994.62*D4.81]
CR = pipe roughness coefficient
D = internal diameter of pipe in m;
h = friction head loss in m;
L = length of pipe in m; and
Q = flow in pipe in m3 / s
Each combination of the pumping head and corresponding pipe diameter should
be capable of supplying the required flow rate over the required distance, and up to
the service reservoir. Smaller pipe diameters will require a higher pumping head to

Interim Report

overcome the increase in head losses, and the other way round. As a result, one
pipe diameter will represent the least cost choice taking into account the initial
costs (capital investment), maintenance costs and the energy costs for pumping.
The total cost, i.e. the capitalized taking both capital and operational cost as the
least and economical, shall be the basis for selecting the most economical pipe
diameter. Therefore, pipe sizing is based upon engineering standards for
acceptable energy (i.e. friction loss = loss of pressure) loss related to the intended
use of the pipe transmission main.
For this analysis, the
calculated costs for
different pipe sizes
shall be are plotted
as in a graph shown
as side. Therefore,
when the capital cost
and

the

operation

cost are added the


concave up curve is
derived. The point of inflection indicates the minimum total Cost. From this point
draw a vertical line that will intersect the X-axis at the most Optimum diameter.
The most economical pipe diameter will tend to be large when energy costs are
high, unit costs of pipe low and capital interest rates low. Nevertheless, it should
not be forgotten that a larger pipe means lower velocity, i.e. potential water quality
problems. As a preliminary estimate, the range of possible most economic
diameters can be selected based on velocities around 1.0-1.8 m/s.
The materials of construction available for construction of large diameter water
transmission main segments include the following alternatives:

Ductile iron pipe (DIP)

Pre-stressed Concrete Cylinder Pipe (PCCP)

Carbon Steel

High Density Polyethylene pipe (HDPE)

Interim Report

Based on historical performance, life expectancy and durability, the use of cement
lined and bitumen lined, and coated DIP and fittings will provide an excellent
conveyance mechanism for all segments of the proposed transmission main for the
portions considered for open cut construction.
In selecting the most appropriate pipe material for the said scheme/project the
following consideration shall be made:

o
o
o

The ground conditions along the pipeline route, mainly conditions such as
traffic overload, proximity to sewer lines, and crowded residential areas.
The cost and local availability of different types of pipe

the design pressure in the distribution system

assess the performance of the pipe material in terms of

Durability,
Resistance to Chemical Attack,
Corrosion and Abrasion. (Suitable external and internal coating shall be made and
if required cathodic protection).
The above aspects are considered in conjunction with the cost of procurement and
installation to provide the most economical solution.
Ductile iron and steel are the strongest pipe materials, making them the best
choice when very high operating pressures are to be expected. These pipes are
almost impossible to tap without special tools and equipment, which is an
advantage in cases where illegal connections pose a real threat for the urban local
body. However, the costs of fittings, valves, etc. increase rapidly for higher pipe
pressure classes and it is therefore often advisable to reduce the maximum internal
pipe pressure through the provision of a pressure reducing valve or break-pressure
tank/ intermediate pumping. A break-pressure tank is generally more reliable than a
pressure reducing valve.
In spite of higher investment costs, ductile iron pipes are a better alternative than
cast iron pipes because they have a longer service life, are lighter and more
flexible and require hardly any maintenance. The pipe is practically corrosion

Interim Report

resistant due to coatings applied inside and outside. Moreover, this pipe material
may be subject to scale bursts when tapped without sufficient skill. Alternative
materials are in this case High-density polyethylene (PE).
High-density polyethylene (PE) is a very suitable pipe material for small-diameter
mains because it can be supplied in coil. The potential of laying this pipe in longer
lengths reduces the number of necessary joints. Particularly in cases where rigid
pipe materials would necessitate a considerable number of special parts such as
elbows and bends, the flexible PE makes for an ideal pipe material. Polyethylene
does not deteriorate when exposed to direct sunlight. Conventional jointing of the
PE-pipes may cause leakage and welding is considered to be a better alternative.
However PE pipes greater than 1600 mm diameter are not commercial available in
the local market, hence is a big constraint in selecting the said pipe material for
commercial use when the diameter requirement is above 1600mm OD.
To summarize, for pipelines of small-diameter (less than 300 mm) PE may
generally be the best alternative unless high working pressures are expected
(above 60 mwc). These pipes can also be used for medium- to large-sized
pipelines (diameters up to 500-600 mm) where lower pressures can be maintained.
However, Cast iron, Ductile iron and Carbon steel are generally only used for largediameter mains and also in cases where very high pressures necessitate their use
in small- or mid-range diameter pipes. Due to heavy weight and lower flexibility, CI
pipes are becoming less advantageous than DI, despite lower prices. Moreover CI
pipes are brittle as compared to DI pipes. Stringent measures that have to be
introduced while handling these pipes involve the prevention of the production and
inhalation of fibre dust (use of special saws cutting under wet conditions, protection
masks for the workers, etc).
Since DI pipe above 1.1m dia are not available commercially, hence all pipes
above 1.1m dia shall be Carbon Steel as per IS 3589 with internal Cement
Concrete Lining (CC) lining in accordance with Is11906 or ISO 4179 and outside
fusion bonded epoxy coating for prevention for corrosion and all pipes below or

Interim Report

equal to 1.1 m shall be DI with internal CC lining and outside fusion bonded epoxy
coating as per IS 8329.
The water pumped from the proposed facility shall be conveyed to existing or new
ground storage tanks for further distribution in the city. This shall be confirmed in
the interim stage, when all surveys are in place.

Water:
Product water

Mild steel (IS 2062,plate fabricated) with mortar lining


inside and fusion bonded epoxy /polyurethane/epoxy
coatings on exposed surface (above 1100 mm
diameter of pipe),
Ductile iron pressure pipes and fittings (Class K9)
shall comply with IS 8329 and IS 9523 (Upto 1100
mm Diameter of pipe).

Interim Report

14.1

Annexure

14.1.1

Pipe sizing and pump capacity for 550 MLD

Capacity

LENGTH
C
TOTAL STATIC HEAD (H)
PIPE MATERIAL (MS)

DESIGN VELOCITY

1
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2
2.1
2.2
2.3
2.4

60000
0.8
6
IS 2062

UNIT

M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S
M/S

550.00

MLD

6.37

M/S

M
M

CALCULATED DIA
( M)

SELECTED
DIA ( M)

ACTUAL
VELOCITY
(M/S)

(Q/C)^1.81

2.846947987
2.714458399
2.598896054
2.496938563
2.406110204
2.324523298
2.250710005
2.183509431
2.121989743
2.06539315
2.013096227
1.964580755
1.919411941
1.877221912
1.837697024

2.9
2.8
2.6
2.5
2.5
2.4
2.3
2.2
2.2
2.1
2.1
2
2
1.9
1.9

0.96
1.03
1.20
1.30
1.30
1.41
1.53
1.67
1.67
1.84
1.84
2.03
2.03
2.25
2.25

42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69
42.69

994.62

994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62
994.62

D^4.81

167.5461
141.524
99.08817
82.05243
82.05243
67.42424
54.94281
44.36629
44.36629
35.47111
35.47111
28.05138
28.05138
21.9182
21.9182

STATIC
HEAD
(H)
6
6
6
6
6
6
6
6
6
6
6
6
6
6
6

15.37
18.20
25.99
31.39
31.39
38.20
46.88
58.05
58.05
72.61
72.61
91.81
91.81
117.51
117.51

REQUIRED PUMP
HEAD INCLUDING
MINOR LOSS (10 %)
IN 'M'

SELECTED
PUMP HEAD (M)

22.91
26.02
34.59
40.53
40.53
48.02
57.56
69.86
69.86
85.87
85.87
107.00
107.00
135.26
135.26

23.00
27.00
35.00
41.00
41.00
49.00
58.00
70.00
70.00
86.00
86.00
107.00
107.00
136.00
136.00

Page 99

HEAD IN
Kg/Cm
2.346
2.754
3.57
4.182
4.182
4.998
5.916
7.14
7.14
8.772
8.772
10.914
10.914
13.872
13.872

Interim Report

S.No.
A

Particulars

Formula/Symbol

Unit

option-i

option- ii

optioniii

optioniv

option-v

optionvi

option-vii

optionviii

optionix

option-x

option-xi

optionxii

optionxiii

optionxiv

optionxv

Pipe sizing based on internal design pressure

A.1

Selected pipe ID

ID

mm

2900

2800

2600

2500

2500

2400

2300

2200

2200

2100

2100

2000

2000

1900

1900

A.2

Pipe OD

OD

mm

2940

2840

2634.4

2534.4

2534.4

2434.4

2332

2232

2232

2132

2132

2032

2032

1928.4

1928.4

A.3

Design Pressure

Dp

kg/sqcm

2.346

2.754

3.57

4.182

4.182

4.998

5.916

7.14

7.14

8.772

8.772

10.914

10.914

13.872

13.872

A.4

Design
Temperature

Td

deg.C

50

50

50

50

50

50

50

50

50

50

50

50

50

50

50

A.5

Allowable stress

Sa

kg/sqcm
(g)

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

A.6

Corrosion
allowance for
carbon steel

cm

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

A.7

Weld joint
efficiency for
ERW pipes

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

A.8

y factor (as per


ASME B31.1 table
104.1.2(A))

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

A.9

Minimum
calculated pipe
thickness

Tc = ((Dp x OD x
0.1)/(2 x (y x Dp +
Sa x J)) + C) x 10

mm

8.19

7.96

7.50

7.28

7.28

7.05

6.82

6.59

6.59

6.37

6.37

6.14

6.14

5.90

5.90

A.10

Mill negative
tolerance for pipe
thickness

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

A.11

Minimum required
wall thickness
(after considering
mill tolerance)

T* = Tc / ( 1 (M/100))

mm

9.36

9.10

8.57

8.31

8.31

8.06

7.79

7.53

7.53

7.28

7.28

7.02

7.02

6.75

6.75

A.12

Selected
thickness of pipe
(based on internal
load thickness
calculation)

Ts

mm

20

20

17.2

17.2

17.2

17.2

16

16

16

16

16

16

16

14.2

14.2

Page 100

Interim Report

Page 101

Interim Report

14.1.2

Pipe sizing and pump capacity for 2 X 275 MLD

Capacity

LENGTH
C
TOTAL STATIC HEAD (H)
PIPE MATERIAL (MS)

60000
0.8
6
IS 2062

275.00
3.18

MLD
M/S

M
M

NOTE
1

SELECTED
DIA ( M)

ACTUAL
VELOCITY
(M/S)

(Q/C)^1.81

994.62

D^4.81

STATIC
HEAD (H)

TWO PIPES EACH ACCOMODATING 275 MLD , THE ENTIRE SYSTEM (2 x 275 MLD ) COMPRISING
PUMPS. REFER SCHEMATIC DIAGRAM PROVIDED IN THE SHEET.

4W + 2S

REQUIRED PUMP
HEAD INCLUDING
MINOR LOSS (10 %)
IN 'M'

SELECTED
PUMP
HEAD (M)

HEAD
IN
Kg/Cm

DESIGN VELOCITY

UNIT

CALCULATED
DIA ( M)

M/S

2.013096227

2.1

0.92

12.18

994.62

35.47111

20.71

28.78

29.00

2.958

1.1

M/S

1.919411941

1.01

12.18

994.62 28.05138

26.18

34.80

35.00

3.57

1.2

M/S

1.837697024

1.9

1.12

12.18

994.62

21.9182

33.51

42.86

43.00

4.386

1.3

M/S

1.76560219

1.8

1.25

12.18

994.62 16.89899

43.47

53.81

54.00

5.508

1.4

M/S

1.701376842

1.8

1.25

12.18

994.62 16.89899

43.47

53.81

54.00

5.508

1.5

M/S

1.643686187

1.7

1.40

12.18

994.62 12.83687

57.22

68.94

69.00

7.038

1.6

M/S

1.591492307

1.6

1.58

12.18

994.62 9.589967

76.59

90.25

91.00

9.282

1.7

M/S

1.543974326

1.6

1.58

12.18

994.62 9.589967

76.59

90.25

91.00

9.282

1.8

M/S

1.500473337

1.6

1.58

12.18

994.62 9.589967

76.59

90.25

91.00

9.282

1.9

M/S

1.460453502

1.5

1.80

12.18

994.62 7.030706

104.47

120.92

121.00

12.342

M/S

1.423473994

1.5

1.80

12.18

994.62 7.030706

104.47

120.92

121.00

12.342

2.1

M/S

1.389168374

1.4

2.07

12.18

994.62 5.045171

145.59

166.15

167.00

17.034

2.2

M/S

1.3572292

1.4

2.07

12.18

994.62 5.045171

145.59

166.15

167.00

17.034

2.3

M/S

1.327396343

1.4

2.07

12.18

994.62 5.045171

145.59

166.15

167.00

17.034

2.4

M/S

1.299448027

1.3

2.40

12.18

994.62 3.532381

207.94

234.73

235.00

23.97

Page 102

Interim Report
S.No.
A

Particulars

Formula/Symbol

Unit

option-i

option- ii

optioniii

optioniv

option-v

optionvi

optionvii

optionviii

optionix

optionx

optionxi

optionxii

optionxiii

optionxiv

optionxv

Pipe sizing based on internal design pressure

A.1

Selected pipe ID

ID

mm

2100

2000

1900

1800

1800

1700

1600

1600

1600

1500

1500

1400

1400

1400

1300

A.2

Pipe OD

OD

mm

2140

2040

1934.4

1834.4

1834.4

1734.4

1632

1632

1632

1532

1532

1432

1432

1428.4

1328.4

A.3

Design Pressure

Dp

kg/sqcm

2.958

3.57

4.386

5.508

5.508

7.038

9.282

9.282

9.282

12.342

12.342

17.034

17.034

17.034

23.97

A.4

Design
Temperature

Td

deg.C

50

50

50

50

50

50

50

50

50

50

50

50

50

50

50

A.5

Allowable stress

Sa

kg/sqcm
(g)

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

cm

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

A.6
A.7

Corrosion
allowance for
carbon steel
Weld joint
efficiency for
ERW pipes

A.8

y factor (as per


ASME B31.1
table 104.1.2(A))

A.9

Minimum
calculated pipe
thickness

Tc = ((Dp x OD x
0.1)/(2 x (y x Dp
+ Sa x J)) + C) x
10

mm

6.39

6.17

5.93

5.71

5.71

5.48

5.25

5.25

5.25

5.02

5.02

4.79

4.79

4.78

4.55

A.10

Mill negative
tolerance for
pipe thickness

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

T* = Tc / ( 1 (M/100))

mm

7.31

7.05

6.78

6.52

6.52

6.26

6.00

6.00

6.00

5.74

5.74

5.48

5.48

5.47

5.20

Ts

mm

20

20

17.2

17.2

17.2

17.2

16

16

16

16

16

16

16

14.2

14.2

A.11

A.12

Minimum
required wall
thickness (after
considering mill
tolerance)
Selected
thickness of
pipe (based on
internal load
thickness
calculation)

Page 103

Interim Report

14.1.3

Pipe size and pump capacity for 400 MLD

Page 104

Interim Report

Capacity

LENGTH
C
TOTAL STATIC HEAD
(H)

PIPE MATERIAL
(MS)

60000
0.8

M
-

400.00
4.63

MLD
M/S

IS 2062

(Q/C)^1.81

994.62

D^4.81

STATIC
HEAD
(H)

REQUIRED
PUMP HEAD
INCLUDING
MINOR LOSS
(10 %) IN 'M'

SELECTED
PUMP
HEAD (M)

DESIGN VELOCITY

UNIT

CALCULATED DIA
( M)

SELECTED
DIA ( M)

ACTUAL
VELOCITY
(M/S)

M/S

2.427885401

2.5

0.94

23.99

994.62

82.052429

17.64

25.40

26.00

2.652

1.1

M/S

2.314897901

2.4

1.02

23.99

994.62

67.424242

21.46

29.61

30.00

3.06

1.2

M/S

2.216346002

2.3

1.11

23.99

994.62

54.942813

26.34

34.97

35.00

3.57

1.3

M/S

2.129396361

2.2

1.22

23.99

994.62

44.366288

32.62

41.88

42.00

4.284

1.4

M/S

2.051937677

2.1

1.34

23.99

994.62

35.471112

40.80

50.88

51.00

5.202

1.5

M/S

1.982360129

1.47

23.99

994.62

28.051383

51.59

62.75

63.00

6.426

1.6

M/S

1.919411941

1.47

23.99

994.62

28.051383

51.59

62.75

63.00

6.426

1.7

M/S

1.8621031

1.9

1.63

23.99

994.62

21.918197

66.03

78.63

79.00

8.058

1.8

M/S

1.809638933

1.9

1.63

23.99

994.62

21.918197

66.03

78.63

79.00

8.058

1.9

M/S

1.761373197

1.8

1.82

23.99

994.62

16.89899

85.64

100.20

101.00

10.302

M/S

1.716774231

1.8

1.82

23.99

994.62

16.89899

85.64

100.20

101.00

10.302

2.1

M/S

1.675400097

1.7

2.04

23.99

994.62

12.836875

112.74

130.01

131.00

13.362

2.2

M/S

1.636880004

1.7

2.04

23.99

994.62

12.836875

112.74

130.01

131.00

13.362

2.3

M/S

1.600900225

1.7

2.04

23.99

994.62

12.836875

112.74

130.01

131.00

13.362

2.4

M/S

1.567193288

1.6

2.30

23.99

994.62

9.5899671

150.91

172.00

173.00

17.646

HEAD IN
Kg/Cm

Page 105

Interim Report

S.No.
A

Particulars

Formula/Symbol

Unit

option-i

option- ii

optioniii

optioniv

option-v

optionvi

option-vii

optionviii

optionix

option-x

option-xi

option-xii

optionxiii

optionxiv

option-xv

Pipe sizing based on internal design pressure

A.1

Selected pipe ID

ID

mm

2500

2400

2300

2200

2100

2000

2000

1900

1900

1800

1800

1700

1700

1700

1600

A.2

Pipe OD

OD

mm

2534.4

2434.4

2334.4

2232

2132

2032

2032

1932

1932

1832

1832

1728.4

1728.4

1728.4

1628.4

A.3

Design Pressure

Dp

kg/sqcm

2.652

3.06

3.57

4.284

5.202

6.426

6.426

8.058

8.058

10.302

10.302

13.362

13.362

13.362

17.646

A.4

Design
Temperature

Td

deg.C

50

50

50

50

50

50

50

50

50

50

50

50

50

50

50

A.5

Allowable stress

Sa

kg/sqcm
(g)

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

cm

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

A.6
A.7

Corrosion
allowance for
carbon steel
Weld joint
efficiency for ERW
pipes

A.8

y factor (as per


ASME B31.1 table
104.1.2(A))

A.9

Minimum
calculated pipe
thickness

Tc = ((Dp x OD x
0.1)/(2 x (y x Dp
+ Sa x J)) + C) x
10

mm

7.28

7.05

6.83

6.60

6.37

6.15

6.15

5.92

5.92

5.69

5.69

5.46

5.46

5.46

5.23

A.10

Mill negative
tolerance for pipe
thickness

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

A.11

Minimum required
wall thickness
(after considering
mill tolerance)

T* = Tc / ( 1 (M/100))

mm

8.32

8.06

7.80

7.54

7.28

7.02

7.02

6.77

6.77

6.51

6.51

6.24

6.24

6.24

5.98

A.12

Selected thickness
of pipe (based on
internal load
thickness
calculation)

Ts

mm

17.2

17.2

17.2

16

16

16

16

16

16

16

16

14.2

14.2

14.2

14.2

Page 106

Interim Report

Page 107

Interim Report

14.1.4

Pipe size and pump capacity for 150 MLD

Capacity

LENGTH
C
TOTAL STATIC HEAD (H)
PIPE MATERIAL (MS)

DESIGN VELOCITY

60000
0.8

M
-

6
IS 2062

UNIT

CALCULATED DIA
( M)

NOTE
1

SELECTED DIA
( M)

150.00
1.74

MLD
M/S

ONE PIPE ACCOMODATING 150 MLD , THE ENTIRE SYSTEM WILL COMPRISE 3W + 2S PUMPS. REFER
SCHEMATIC DIAGRAM PROVIDED IN THE SHEET.

ACTUAL
VELOCITY
(M/S)

(Q/C)^1.81

994.62

STATIC
HEAD
(H)

D^4.81

REQUIRED
PUMP HEAD
INCLUDING
MINOR LOSS
(10 %) IN 'M'

SELECTED
PUMP
HEAD (M)

M/S

1.486770097

1.5

0.98

4.06

994.62

7.030706299

34.88

44.36

45.00

HEAD IN
Kg/Cm
4.59

1.1

M/S

1.417579666

1.5

0.98

4.06

994.62

7.030706299

34.88

44.36

45.00

4.59

1.2

M/S

1.3572292

1.4

1.13

4.06

994.62

5.045170546

48.60

59.46

60.00

6.12

1.3

M/S

1.303983636

1.4

1.13

4.06

994.62

5.045170546

48.60

59.46

60.00

6.12

1.4

M/S

1.256550073

1.3

1.31

4.06

994.62

3.532380874

69.42

82.36

83.00

8.466

1.5

M/S

1.213942701

1.3

1.31

4.06

994.62

3.532380874

69.42

82.36

83.00

8.466

1.6

M/S

1.175394966

1.2

1.54

4.06

994.62

2.403597776

102.02

118.22

119.00

12.138

1.7

M/S

1.140300611

1.2

1.54

4.06

994.62

2.403597776

102.02

118.22

119.00

12.138

1.8

M/S

1.108173001

1.2

1.54

4.06

994.62

2.403597776

102.02

118.22

119.00

12.138

1.9

M/S

1.078616395

1.1

1.83

4.06

994.62

1.581607864

155.04

176.54

177.00

18.054

M/S

1.051305217

1.1

1.83

4.06

994.62

1.581607864

155.04

176.54

177.00

18.054

2.1

M/S

1.025968838

1.1

1.83

4.06

994.62

1.581607864

155.04

176.54

177.00

18.054

2.2

M/S

1.002380195

1.1

1.83

4.06

994.62

1.581607864

155.04

176.54

177.00

18.054

2.3

M/S

0.98034717

2.21

4.06

994.62

245.21

275.73

276.00

28.152

2.4

M/S

0.959705971

2.21

4.06

994.62

245.21

275.73

276.00

28.152

Page 108

Interim Report

S.No.
A

Particulars

Formula/Symbol

Unit

option-i

optionii

option-iii

option-iv

optionv

optionvi

optionvii

optionviii

optionix

option-x

option-xi

optionxii

optionxiii

optionxiv

optionxv

Pipe sizing based on internal design pressure

A.1

Selected pipe ID

ID

mm

1500

1500

1400

1400

1300

1300

1200

1200

1200

1100

1100

1100

1100

1000

1000

A.2

Pipe OD

OD

mm

1540

1540

1434.4

1434.4

1334.4

1334.4

1232

1232

1232

1132

1132

1132

1132

1028.4

1028.4

A.3

Design Pressure

Dp

kg/sqcm

4.59

4.59

6.12

6.12

8.466

8.466

12.138

12.138

12.138

18.054

18.054

18.054

18.054

28.152

28.152

A.4

Design
Temperature

Td

deg.C

50

50

50

50

50

50

50

50

50

50

50

50

50

50

50

A.5

Allowable stress

Sa

kg/sqcm
(g)

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

1311.43

A.6

Corrosion
allowance for
carbon steel

cm

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

0.16

A.7

Weld joint
efficiency for ERW
pipes

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

0.85

A.8

y factor (as per


ASME B31.1 table
104.1.2(A))

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

0.4

A.9

Minimum
calculated pipe
thickness

Tc = ((Dp x OD x
0.1)/(2 x (y x Dp +
Sa x J)) + C) x 10

mm

5.05

5.05

4.81

4.81

4.58

4.58

4.35

4.35

4.35

4.12

4.12

4.12

4.12

3.88

3.88

A.10

Mill negative
tolerance for pipe
thickness

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

12.5

A.11

Minimum required
wall thickness
(after considering
mill tolerance)

T* = Tc / ( 1 (M/100))

mm

5.77

5.77

5.50

5.50

5.24

5.24

4.97

4.97

4.97

4.71

4.71

4.71

4.71

4.44

4.44

A.12

Selected
thickness of pipe
(based on internal
load thickness
calculation)

Ts

mm

20

20

17.2

17.2

17.2

17.2

16

16

16

16

16

16

16

14.2

14.2

Page 109

Interim Report

Page 110

Interim Report

15.

BENEIFTS OF LONG-TERM O&M CONTRACT


Municipalities/utilities enter into long-term O&M Contract services to achieve a number of
goals, including:

Reducing costs, both capital and O&M costs;

Increasing efficiency;

Enhancing risk management;

Meeting capital investment needs;

Achieving regulatory compliance;

Overcoming lack of local expertise through private-sector experience; and

Saving time.

Some of these objectives may be contradictory. For example, it may not be possible to
extract the greatest financial value and, at the same time, reduce rates. Similarly, it may
not be able to realize significant cost reductions and, at the same time, protect
employees entirely. These choices are predicated on trade-offs, i.e. what can be
exchanged for the guarantees afforded by an O&M agreement. The stability and
continuity of a 10- 15-year contract is quite clear over greater immediate financial gains
possible through a short-term agreement.
The essence of public-private partnerships is the type and breadth of guarantees that are
rare and difficult to achieve under public operation and management. Essentially, the
guarantee aspect translates to an enhanced risk-management advantage. The more
responsibility given to the private sector, the better the risk profile from the government's
viewpoint. Guarantees in long-term O&M contracts often include:

Guaranteed annual operating budgets and costs;

Guaranteed system operations, regulatory compliance, service quality;

Guaranteed construction costs and facility start-up schedules:

Guaranteed customer service and response: and

Guaranteed revenues and revenue collection

Achieving these guarantees and goals often requires a long-term commitment to


improving the quality and efficiency of municipal water service delivery.

Page 111

Interim Report

In addition to direct cost, compliance, performance, and financial benefits, long-term


contracts for water services can enable local leaders to concentrate their energies on other
programs and functions. The day-to-day management of utility systems has become a
burden in many communities, draining professional and management resources that can
be better focused on other municipal and community needs. In addition, partnerships lead
to increased accountability, improved service levels, capital improvements, and additional
benefits to the community.
The basic main of having long term O&M contracts have Cost Saving and Accountability
As long-term agreements shift responsibility for compliance with all regulations to private
providers, many municipal officials feel a "peace of mind" knowing an experienced
private firm is handling operations and complying with regulations. The more discrete
project components that are tied together, the greater degree of accountability can be
achieved. Investor-owned utilities have also been able to provide a higher level of
customer service at a lower cost by integrating customer-service functions such as call-in
centers, billing and collections into parent company systems.
Through the risk-allocation provisions of a partnership, a community assigns to the
private partner financial responsibility for certain cost overruns, non-compliance, missed
schedules, and/or poor revenue collections. tinder continued public operation, the only
recourse is to raise rates, pay fines, incur additional debt, or subsidize operations.
Further, Long-term contracts often result in improved performance and more
efficient services. Private firms make greater up-front investments in advanced
computers and cutting-edge technology, knowing the costs can be amortized over a
longer period.
Private firms can not only generate significant O&M savings, but can also reduce
capital costs between 10 and 15 percent through the design/build/operate (DBO)
approach. Savings can be plowed back into system improvements and other capital
needs.
Long-term contracts can produce other important benefits for the community. Hiring
the existing workforce promotes continuity and helps avoid nasty, divisive labor battles.
Utilities can also enhance local economic development through long-term contracts.

Page 112

Das könnte Ihnen auch gefallen