fundamentals of grit removal and grit system evaluation_part 2_color very good

4/15/2013

1

This Webcast Is Brought To You By:

Smith & Loveless Inc.

The Fundamentals of Grit Removal and Grit System

Evaluation – Part 2

April 18th, 2012

4/15/2013

2

What: Grit Properties, Settling Principles & Characterization

Steven P. Gress, P.E.

October 15, 2011

Presentation Outline

1. Physical Properties of Grit

2. Traditional Assumptions vs. Recent Suggested Grit Understanding

3. Sand Equivalent Size (SES) Parameter

4. Settling Principles

5. Grit Characterization

Grit Book –Eutek Systems, Inc. 2007

4/15/2013

3

Physical Properties of Grit in Wastewater• Traditional Definition

– Settleable, high density inorganic solids

– Settles as discrete silica sand particles with specific gravity of 2.65 (EPA, 1987)

– Defined as particles larger than 210 microns (65 mesh) by the EPA

– Non-putrescible (clean sand in clean water)

– Sand, gravel, minerals, cinders, other heavy solid material

Various Grit Particle Size Units of Measurement

4/15/2013

4

Traditional Assumptions vs. More Recent Suggested Grit Understanding

Grit Happens - INWEA 2010

Traditional Assumptions More Recent Understanding

All silica sand Variety of material types

2.65 specific gravity Range of specific gravity

Particles larger than 210 microns (μm) (65 mesh)

Particles range in size from 50 to 1000 μm

Perfect spheres Variety of shapes

Clean sand in clean water Fats, oils, greases, soap & scum in collection system

Quiescent basin for discrete particle settling

Basins not quiescent

Specific Gravity Values of Various Grit Materials


Specific Gravity of Various Materials

Quartz Sand 1.2 Earth 1.4

Limestone 1.55 Granite 1.65

Clay 1.8 Red Brick 1.9

Sand, wet 1.92 Gravel 2.0

Asphalt 2.2 Concrete 2.4

4/15/2013

5

Distribution of Grit at WWTPs

Fine Grit Removal Challenges in WWTPs - WEFTEC 2004

Suggested Sand Equivalent Size (SES) Parameter

• Wastewater becomes coated with lower-density FOG– Reduces specific gravity

– Reduces particle settling velocity

• SES Definition– Clean sand particle size with the

same settling velocity as the more buoyant grit particle

– Determined by using a wet sieve analysis


4/15/2013

6

Grit Size vs. SES Size Comparison

Grit Book - Eutek Systems, Inc. 2007

Other Recent Grit Opinions

• No peer reviewed, industry accepted procedure or Standard Method for SES

• Some argue that the FOG coating is thin and effective SG is affected very little (Rippon, Higgins, and Mrkvicka, 2010)

• Others argue that blackened grit particles contain a thin iron oxide or iron sulfide coating that does not radically alter SG

4/15/2013

7

Settling Principles• Discrete particle settling theory

– Newton’s Lawvp(t) = [(4g/3Cd)((ρp – ρw)/ρw)dp]0.5 or

vp(t) ≈ [(4g/3Cd)(sgp – 1)dp]0.5

Where:

• vp(t) = terminal velocity of particle (m/s)

• g = gravity acceleration (9.81 m/s2)

• ρp = density of particle (kg/m3)

• ρw = density of water (kg/m3)

• Cd = drag coefficient (unitless)

• sgp = specific gravity of particle

• dp = diameter of particle (m)

Wastewater Engr. - Trt. & Reuse, 4th Edition

Particle Settling Theory Cont.• Coefficient of Drag Cd takes on different

values depending if flow is laminar or turbulent:– Laminar (NR < 1)

– Transitional (NR = 1 to 2000)

– Turbulent (NR > 2000)

Cd = 24/NR + 3/(NR)0.5 + 0.34

– Reynolds number is defined as:

NR = vpdpρw/μ = vpdp/(kv)

Where: μ = dynamic viscosity (N*s/m2)

kv = kinematic viscosity (m2/s)


4/15/2013

8

Cd As a Function of NR


Particle Settling Theory Cont.

• Newton’s law with proposed modification for non spherical particles:

vp(t) = [(4g/3CdΦ)((ρp – ρw )/ρw)dp]0.5

or

vp(t) ≈ [(4g/3CdΦ)(sgp – 1)dp]0.5

Where:

Φ = shape factor (unitless)

and

• Φ = 1.0 for spheres

• Φ = 2.0 for sand grains


4/15/2013

9


• Settling in Laminar Region (NR < 1)– For NR < 1, viscosity is the predominant force

governing the settling process and the first term in the Cd equation predominates

– Assuming spherical particles, substitution of the first term in the Cd equation into Newton’s law yields Stokes’ law where:vp = g(ρp – ρw)/dp

2/18μ ≈ g(sgp – 1)/dp2/18kv



• Settling in Transition Region (NR = 1 to 2000)

– The complete form of the Cd equation must be used to determine settling velocity

– Because of the nature of the Cd equation, finding the settling velocity is an interactive process

– Utilize NR definition with shape factor to check NR for an assumed Cd value

NR = Φvpdpρw/μ = Φvpdp/(kv)


4/15/2013

10

Settling Velocities in Laminar & Transition Regions at 20°C


Particle Settling Theory Cont.• Settling in Turbulent Region (NR > 2000)

– Inertial forces are predominant and the effect of the first two terms in the Cd

equation is reduced

– Use Cd =0.4

– Resulting settling velocity equation:

vp = [3.33g((ρp – ρw)/ρw)dp]0.5

or

vp = [3.33g(sgp – 1)dp]0.5


4/15/2013

11

Discrete Particle Settling• Usual design procedure is to select a

particle with a terminal velocity vc

– All particles that have a terminal velocity greater than vc will be removed

– For continuous-flow sedimentation• vc = basin depth/detention time

• In actual practice, design factors must be adjusted to allow for effects of inlet and outlet turbulence, short circuiting, solids storage, and velocity gradients due to operation of solids-removal equipment


Definition Sketch for Ideal Discrete Particle Settling


4/15/2013

12

Discrete Particle Settling Cont.• Rate at which clarified water is produced:

Q= Avc or vc = Q/A

Where:– Q = flow rate (m3/s or ft3/s)

– A = surface area of basin (m2 or ft2)

– vc = terminal particle settling velocity (m/s or ft/s)

• Particles with a settling velocity < vc will be removed in the ratio:Xr = vp/vc, where Xr = fraction of particles with settling

velocity vp that are removed


Discrete Particle Settling Cont.

• The total fraction of particles removed: Fraction removed = [∑n

i=1 (vni /vc)(ni)]/∑ni=1 ni

Where:

– vni = average velocity of particles in ith velocity range

– ni = number of particles in the ith velocity range

– When vni /vc is greater than 1.0, use a value of 1.0 because all particles in ith velocity range are removed


4/15/2013

13

Gravity Separation In An Accelerated Flow Field

• Because the top of the separator is enclosed, the rotating flow creates a free vortex within the separator

• The most important characteristic is:

Vr = CWhere:

– V = tangential velocity (m/s or ft/s)

– r = radius (m or ft)

– C = moment constant


Gravity Separation In An Accelerated Flow Field Cont.• As the radius is reduced, the tangential

velocity inversely increases

• Centrifugal force on a particle in this flow field is equal to the square of the velocity divided by the radius– A fivefold decrease in the radius results in a 125-

fold increase in centrifugal force

– Centrifugal acceleration is often expressed in terms of number of “gravities” = G = V2/gr

• g = 32.2 ft/s2


4/15/2013

14

Induced Vortex


Wall velocity is greater than at the centerIncreasing performance as flows decrease (lower SLR)Headloss typically 6-12” for hydraulic induced vortex

Gravity Separation In An Accelerated Flow Field Cont.• Induced vortex type separator typically is a large

diameter cylinder with a conical bottom– Typically has a higher headloss than mechanical vortex type

systems

• Because of high centrifugal forces near the discharge port, grit particles are retained while organic particles are discharged

• Large organic particles typically have a higher drag forces than smaller grit particles and tend to move with the fluid and are transported out of the separator– Organic particles that may settle usually consist of oil and

grease attached to grit particles


4/15/2013

15

Grit Characterization• Methods of grit sampling and analysis

– Sand Equivalent Size (SES) determination of collected influent grit sample

– Dry sieve analysis of collected influent grit sample

– Sampling and characterization of primary sludge

– Sampling and characterization of grit that has accumulated in aeration tanks

• Two inherent difficulties with grit testing (MOP-8)– Uneven distribution of grit in both the influent and

effluent

– Erratic loading patterns associated with grit only being carried into a plant during high flow

Sand Equivalent Size Determination • Sampling Procedures

– Utilizes a circular grit settling device with a conical bottom

– Utilizes a vertically integrated slot sampler (VISS) to sample equally over the entire channel depth

– Utilizes a high volume trash pump placed inside VISS centrally in a vertical upward flow

• Pumps to the settling device

• Trash pump velocity is set to match the velocity across the slot in the VISS

• Care must be taken to avoid pumping from the channel bottom where deposited gravel will be sampled


4/15/2013

16

Sand Equivalent Size Determination Sampling Equipment


Sand Equivalent Size Determination Grit Analysis

• Wet sieve grit sample immediately following collection (time consuming process)– Sieve from ¼” to 50 μm in 10 size intervals in 45 min.

• Measure the settling velocity distribution of each sieve size fraction in a modified Imhoff cone

• Calculate grit SES values

• Plot the SES distribution for each size fraction– Most of the entering grit will have settling velocities <

225 μm sand

• Measure the ash weight for each size interval


4/15/2013

17

Sand Equivalent Size Determination Grit Analysis Cont.• From SES distributions for each size range

and weight distribution, calculate SES weight distribution of influent grit

• Summarize results on companion plots


Dry Sieve Analysis of Collected Influent Grit Sample

• Obtain grit samples from multiple points across channel (cross-channel sampling)– Utilize narrow or slender sampling equipment to

limit disruption of flow velocity

– Label all collected samples by the time of day, collection location, and channel velocity

• Dry–sieving grit samples entails burning samples in a muffle furnace to remove organic material particles prior to sieving– Historically practiced method

– Avoids capillary action between wet fine screens

2008 Smith & Loveless, Bulletin 9503-14

4/15/2013

18

Dry Sieve Analysis Concerns• Potential problems

– The attached surface active agents (FOG & soap) are removed and their buoyant effect on grit is destroyed


– Size distributions of grit before and after ashing in a muffle furnace may vary due to fracturing of some inorganic materials such as limestone

Sampling & Characterization of Primary Sludge

• For Plants with primary clarifiers it is recommended in WEF MOP-8 to sample and classify grit quantities in Primary Sludge– Considered to be a reliable method in MOP-8

– Dry sieve analysis performed

– Analysis may not accurately reflect influent grit• When grit is combined with sludge there is a significant

settling velocity reduction for fine grit

• Actual grit quantities may be larger and grit sizes smaller than analysis results indicate

4/15/2013

19

Sampling & Characterization of Grit Accumulated in Aeration Basins

• Sample and characterize grit (particle size & density) that has accumulated in aeration basins– Accumulated grit is from an extended period &

overcomes problems with a brief sampling program

– Avoids problem with getting a representative sample in a channel due to:

• Improper sampling location

• Turbulence caused by sampling equipment

– Analyzed sample represents a wide range of influent flow and weather conditions

– Provides information on desirable design particle size to avoid grit deposition

Summary & Conclusions• Design based on traditional grit assumptions may result

in low removal efficiency

• Specific gravity of particles and size distribution of grit can vary widely

• FOG coating of grit particles may impact grit settling vel.– Design conservatively

• Grit settling is based on discrete particle settling theory or gravity separation in an accelerated flow field– Avoid turbulent region for discrete particle settling

• Confidence in short term grit characterization studies should be tempered (need to sample during high flows)– May want to sample grit where it accumulates in a WWTP

• Standard sampling and analysis methods are desirable

4/15/2013

20

Questions & Comments

For more information…

Steven P. Gress, P.E.

Technical Associate(317) 500-4221

[email protected]

Case Studies for Analyzing Aerated Grit Basins Using CFD

Marie Burbano, Ph.D., P.E., BCEE, Carrie Knatz, P.E.

4/15/2013

21

Agenda• Introduction

• Strategy For Analyzing Aerated Grit Tanks Using CFD (Computational Fluid Dynamics)

• Case Studies and Results– Greater Augusta Utilities District (GAUD), Augusta, Maine

– Onondaga County, Metropolitan Syracuse Wastewater Treatment Plant (WWTP), Syracuse, New York

– Water Pollution Control Facility (WPCF), City of Gloucester, Massachusetts

• Lessons Learned

• Closing

Introduction

• Computational Fluid Dynamics (CFD)

• CFD predicts hydraulic performance inside a structure using principles of fluid flow known – Reynolds-averaged Navier-Stokes (RANS) equations

– Conservation of mass, momentum and energy

• RANS equations solved to predict hydraulic information– Velocity and pressure

– At a finite number of discrete points within the fluid domain

• Model represents the real-scale geometry

• Analysis can include simulations of different flow conditions

• Results can be extracted from anywhere within the fluid domain

4/15/2013

22

Strategy for Analyzing Aerated Grit Tanks Using CFD

1. Perform physical evaluation of structure– In person

– Photos

– As-builts

2. Compare facility to MOP-8 parameters

3. Identify boundaries of model

4. Establish potential modifications up front

Strategy For CFD Analysis (continued)

5. Input data from as-built drawings, flow rate, and water surface elevation to build model geometry

6. Select size of mesh– Tighter mesh used in areas of concerns

– Larger mesh size used in other areas to speed up simulations

7. Run baseline conditions

8. Run simulations of modifications

9. Review output results (graphics) identify flow patterns and potential disruptions in roll pattern to make recommendations

4/15/2013

23

Greater Augusta Utilities District (GAUD)

• Combined collection system– Average DWF = 5 mgd

– Peak WWF to the plant = 43-44 mgd

– CSO control improvements under design

• WWTF improvements part the program– Wet weather treatment issues

– Low-cost improvements sought –“no new tanks”

Secondary Clarifiers

AerationTanks

PrimaryClarifiers Grit Tanks

© 2009 Google

Grit Chambers

• 2 aerated grit chambers

• Significant, problematic grit carryover– Noticed when flow reaches 5-6

mgd per grit chamber

– Peaks reach 22 mgd per grit chamber

– Collection system improvements to result in peak flow of 18 mgd per grit chamber

2 Grit Tanks

Large baffleat end ofeach tank

Grit collector

4/15/2013

24

GAUD Grit Chambers

Traditional Design Criteria

(Previous MOP 8)

Recent Design Criteria

(New MOP 8)

Width:Depth Ratio 1.25:1 1:1 to 5:1 0.8:1 to 1:1

Length:Width Ratio 1.7:1 3.5:1 to 5:1 3:1 to 8:1

Aeration Rate 2.5 to 6.5 cfm/ft1.5 to 4 cfm/1000 cf

3 to 8 cfm/ft10 to 20 cfm/1000 cf

HRT at peak flow 3 min (@43 mgd) 2 to 5 min 3 to 10 min

Side Water Depth 12.75 ft 12 to 16 ft

Floor slope 25 percent >30 percent

Aerated Grit Chamber Design Criteria

Conclusions of Preliminary Evaluation• Grit chamber design criteria confirm observed

performance

• Total volume of grit chambers limits likely improvement at peak flow

• Improvement should be possible at some intermediate lower flow

• CFD would be useful in identifying and optimizing low-cost improvements

4/15/2013

25

CFD Scenarios

1. Existing grit chamber

2. Existing baffle removed

3. New “lateral” baffles

4. New “longitudinal” baffle

5. Re-sloped floor

3 flows: 6 (low), 12 (avg), 22 (peak) mgd

(1) Proposednew

“lateral”baffles

Proposedfloor

resloping

Proposed“longitudinal”

baffle

CFD Geometry and Model Setup• Gambit (version 2.4.6)

preprocessor– Mesh generated

– Nearly 1,000,000 cells

• Fluent version 6.3– Steady-state analysis

– Reynolds-averaged, Navier-Stokes equations describing flux of mass and momentum

– Standard k-epsilon turbulence model

• Aeration– Discrete phase model (DPM)

– DPM injects air particles to affect primary flow pattern

4/15/2013

26

Existing baffle causes upward

velocity in area where grit

has been collected and

conveyed.

Pathlines colored by velocity

Existing BafflePeak flow – 22mgd

No upward velocity at

effluent end; spiral pattern

continues.

Pathlines colored by velocity

Remove Existing BafflePeak flow – 22mgd

4/15/2013

27

Core of very low V

Higher V around the core

Higher V above collection screw upstream of weir

Existing baffle

Core less pronounced

Lower V above

collection screw

Existing baffle removed

Core less pronounced, Similar to

no-baffle results

Proposed lateral baffles


Proposed longitudinal baffle

Comparison – Peak flow (22 mgd)

Pronounced core of very low V

Higher V around the

core

Upflow V approaching

weir

Existing baffle

Overall more consistent V

Existing baffle removed

Core varies but exists throughout

tank length

Overall very similar to the no-baffle

results

Proposed lateral baffles


Higher V around the core

Proposed longitudinal baffle

Comparison – Peak flow (22 mgd)

4/15/2013

28

Recommended Improvements

• Improvements completed – observed improved grit removal

(1) Removeexisting baffle

(2) Addproposed

“longitudinal”baffle

Onondaga County, Metropolitan Syracuse WWTP• Combined Sewer

• 2 Screening and Grit Removal Facilities– New Screening & Grit (NS&G)

– Existing Screening & Grit (ES&G)

• Flows– Peak Flow: 240 MGD

– Design Flow: 84 MGD

– Average Flow: 65 MGD (2010)

– Tertiary Treatment: <126 MGD

NSG

ESG

PS

Headworks

4/15/2013

29

ES&G Aerated Grit Chambers

• Flow enters at the bottom of the tank perpendicular to flow direction

• 2 intermediate baffles per tank

• Grit removal from influent end

• 40’ long, 13’ wide, 12’ deep

• 19 diffusers, 160 cfm, air velocity = 5 ft/sec

Failure due to

grit build up

NS&G Aerated Grit Chambers• Flow enters near top of tank

perpendicular to flow direction

• 2 intermediate baffle walls

• Grit removal from center (2 screw conveyors per tank)

• Two 104’ long, 25’ wide, 13’ deep chambers

• 100 diffusers, 1500 cfm, air velocity = 11 fps

4/15/2013

30

Grit Removal Project Goals

• ES&G and NS&G– Optimize grit removal by improving hydraulic flow pattern to

induce vertical roll pattern

– Distribute grit along length of tank

• NS&G– Evenly divide grit between grit chambers 1 and 2

• Create basis for design by using CFD– CFD used as a tool to help maximize value by spending dollars

on most effective modifications

– Modeled existing configuration as a baseline to be able to determine level of improvement achieved by modifications

Comparison of Grit Chambers to MOP-8 Design Criteria

Criteria MOP-8- 5th Ed ES&G NS&G

Width : Depth Ratio 0.8 to 1 1.12 1.92

Depth 12 to 16 feet 11.9 13.0

Length : Width Ratio 3:1 to 8:1 3:1 4:1

Air Rates 3 to 8 cfm/ft 4 cfm/ft 14.2 cfm/ft

CFD Modeling of Grit Improvements

4/15/2013

31

ES&G: Model Simulation Results for Baseline Conditions

ES&G: Proposed Model Simulations

• Once baseline simulations were complete, recommendations for proposed modifications were selected– Remove intermediate baffles

– Increase air flow

– Modify influent “doghouse” baffle

4/15/2013

32

ES&G Comparison of Baseline and ProposedES&G Baseline ES&G Proposed

Grit Tracer Study

ES&G – Proposed

ES&G – Baseline

4/15/2013

33

NS&G1 and NS&G2: Baseline Conditions

NS&G 1 NS&G 2

Future Work

• ES&G: evaluate longitudinal baffle possibly

• NS&G 1: evaluate design of baffles to straighten influent flow

• NS&G 2: likely similar modifications as NSG 1, other than how flow is straightened

4/15/2013

34

Gloucester Water Pollution Control Facility (WPCF)

• Purpose: Assess overall hydraulic performance with respect to potential for grit re-suspension

• CFD model (Fluent version 12.1) – Single-phase, steady state analysis

– Simulated aeration by introducing air bubbles with discrete-phase model (DPM)

• WPCF has two identical aeration grit tanks operating in parallel only one tank was modeled

• Two flow conditions were simulated: – Case 1: 15 mgd with one tank in service

– Case 2: 15 mgd with two tanks in service (assume equal flow split between the two tanks)

Gloucester WPCF Results

Case 1: One tank for full 15 mgd flow

Case 2: Two tanks in service, split 15 mgd flow

4/15/2013

35

CFD Lessons Learned

• Develop potential modifications early and build into the geometry of the model, but be prepared for additional effort if baseline simulation shows something unexpected

• Workshops/collaboration important – team approach

• If something is going to be changed in near future, modify the geometry and include it in baseline simulation as a given

• If funding available, model at different flow rates

• Review overall impacts of changes (upstream/downstream hydraulics for example)

Closing• Acknowledgements

– GAUD• Dan Bisson (CDM), Bill McConnell (CDM), Brian

Tarbuck (GAUD)

– Augusta County• Dan Jean (GAUD), Jim Renk (GAUD), Nancy

Vigneault (CDM)

– Gloucester• Mike Kostur (CDM)

• Questions?

4/15/2013

36

Vortex Case StudiesFull Scale and Pilot Scale

by

Brian F. McNamara

Grit Removal Efficiency

Grit Unit Grit Effluent

Final Product

Grit Influent

Sample Location

%Grit Efficiency = Grit Inf – Grit Eff x 100Grit Inf

4/15/2013

37

Vertical Sampler Technique

This technique was employed for both full scale and pilot scale

studies

Typical pump rate approximately 120 gpm

Vertical SamplerGathers a representative sample from the entire vertical water column

Slot width function of stream velocity, channel depth, pump velocity, & cross sectional area of pump suction

4/15/2013

38

50-micron Grit Settler

Sampling begins as diurnal flows raise up and continue thru the peak of the afternoon. Continuous composite for 6 hours.

Post-Sampling Activities

• Decant sample from settler

• Rinse loose floatable organics from sample

• Drain off liquid

• Volume & Weight of total sample recorded

• Wet-sieve

4/15/2013

39

Wet-sieve System

Fixed Solids Determination

Flow weighted calculations for quantity or grit per MGD

Grit vs. Grit

Pilot Evaluation Comparing

Multi-Tray

verses

Structured

Vortex Grit Removal Technologies

4/15/2013

40

Multiple Tray Technology

Multiple Tray Features

• Full Scale

• Two Units

• 18 MGD each (68 ML/d)

• 9 Trays

– 12 foot dia (3.66 m)

– 1062 ft2 (98.7 m2)

• Plastic & Stainless Steel

• Two foot head loss (60 cm)

• Cut Point 100 µm @ 2.65 sg

4/15/2013

41

Multiple Tray Pilot Unit

Two Trays 4 ft diameter (1.22 m)Optimum Performance: Feed Rate170 gpm Cut point : 75 µm @ 2.65 sg

Structured Vortex Technology

4/15/2013

42

Structured Vortex Features

• Full Scale

• Two Units

• 18 MGD each (68 ML/d)

• 22 foot diameter (6.7 m)

• Concrete & Stainless Steel

• Less than one foot head loss (30.5 cm)

• Cut point 150 µm @ 2.6 sg

Structured Vortex Pilot Unit

4 ft dia (1.22m)Optimum Performance: Feed Rate180 gpm Cut point : 106 µm @ 2.65 sg

4/15/2013

43

Test Protocol

• Simultaneous Parallel testing

• Same day testing

• Use native grit from plant RWI

• Comparable flow rates

• Applicable full scale flow rates

• Same test procedures

• Same analysis

• One laboratory

Scale Up Test Protocol

4/15/2013

44

Pilot Testing

Dual Vertical Samplers at RWI

4/15/2013

45

Pilots with Sampling Equipment

Testing Problems

•Ultrasonic Flow Meters could not perform

•Feed rates determined by fill test and adjusting throttle on pump motor

•Rags fouled pump suctions•Bar screens placed in hand

•Could not achieve pump rates of 300 gpm (1136 Lpm) or the full scale equivalent of 18 MGD (68 LPD)

•RWI channel 22 ft deep•Trash pumps too small•Rented pumps unable to perform

4/15/2013

46

10 MGD ResultsMultiple Tray39 ML/d (10.4 MGD) December 17, 2007

Structured Vortex 40 ML/d (10.5 MGD) December 17, 2007

15 MGD Results

Structured

Vortex

56 ML/d

(14.8 MGD)

December 18,

2007

Multiple Tray 57 ML/d (15.1 MGD) December 18, 2007

4/15/2013

47

Pilot Unit Removal Efficiencies

Conclusion

• Foot Print Equal

• Price Comparable– $1.3 Million for 2 Multiple Tray Units

– $1.2 Million for 2 Vortex Units

• Head loss greater for multiple tray unit

• Multiple tray unit exhibited higher removal rates at higher flows

4/15/2013

48

True Grit

Full-Scale Performance Assessment At Three

Mechanical Vortex Grit Removal Treatment Plants

*Specs indicate both 30 MGD and 50 MGD**All systems reference grit at 2.65 SG

Note: All removal processes refined the grit using cyclones & classifiers

Evaluated Grit Removal Processes

4/15/2013

49

VIP Forced Vortex 20 ft dia

Plant Design 40 MGD with Collection system composed of both gravity and force mains

Vortex Grit Effluent

Grit Refinement

Final Product

RWI

Plant Recycles

Plant Drains

Grit Influent

Recycle


54” vertical rise

Preceded by RWI wet well

29 MGD

RWI = 2.8 fps

Grit Inf = 2.3 fps

Grit Eff = 3.5 fps

4/15/2013

50

VIP RWI Forced Vortex 20 ft dia

Grit Influent and

Grit Effluent


4/15/2013

51

Grit in primary biosolids

Premature wear on centrifuges

Grit volume taking up valuable incinerator space

Plant HistoryVIP Forced Vortex 20 ft dia

Plant HistoryVIP Forced Vortex 20 ft dia

RWI flow and Incinerator feed belt totalizer during a hydraulic event.

4/15/2013

52

Concentration of Grit VIP Force Vortex 20 ft dia on May 20, 2007Sunday

Average hourly flow during sampling was approximately 30 MGD

Specification requires 65% of 100 mesh (150 μm) at a maximum design flow of 26.7 MGD and 2.65 sg

Virginia Initiative PlantVortex 20 ft Dia Efficiency Results

4/15/2013

53

System ConclusionsVIP Forced Vortex 20 ft dia

- 70% of RWI grit was over 200 microns

- > 300 micron grit shape flat, settled slowly

- Current system operation does not adequately protect downstream equipment

- Note: Plant was advised to use 2 vortexes if flows over 26 MGD

CETP Forced Vortex 24 ft dia

Plant Design 24 MGD with a collection system composed of force mains

4/15/2013

54


Grit Refinement

Final Product

RWI

Plant Recycles

Plant Drains

Grit Influent

Recycle


Force main collection system. Predominantly urban.

23 MGD

RWI = 2.0 fps

Grit Inf = 4.0 fps

Grit Eff = 3.0 fps

Parshall Flume

Sample Location

CETP Forced Vortex 24 ft dia RWI & Grit Influent

23 MGD RWI = 2 fps Grit Inf = 4 fps Grit Eff = 3 fps

4/15/2013

55


Grit Effluent

Plant HistoryCETP Forced Vortex 24 ft dia

Prior to forced vortex units, plant used square grit chambers or Detritors.

2005 Detritors removed 24,570 ft3 grit

2007 Forced vortex & hydro-cyclones removed 2,034 ft3

4/15/2013

56

Plant Grit ExperienceCETP Forced Vortex 24 ft dia

Grit Build up in the aeration tanks

Lost Secondary Clarifier due to grit build up

Removed grit from contact tanks

Removed grit in scum concentrator

Concentration of Grit CETP Forced Vortex 24 ft diaRWI, Grit Influent, Grit EffluentThursday May 17, 2007

4/15/2013

57

CETP Forced Vortex 24 ft dia Removal Efficiency

At 30 MGD, manufacturer specifies 95% removal of 100-mesh with 2.65 sg

Average hourly flow during sampling was approximately 20 MGD/hour

System ConclusionsCETP Forced Vortex 24 ft dia

- +80% of RWI grit was over 200 microns

- > 300 micron grit shape flat, settled slowly

- System did not protect downstream equipment

4/15/2013

58

Nansemond Treatment Plant24 ft Vortex

Plant Design 30 MGD

with a collection system composed of force mains


Grit Refinement

Final Product

RWI

Plant Recycles

Plant Drains

Grit Influent

Recycle

NTP Forced Vortex 24 ft dia

48” vertical rise

22.5 MGD

RWI = 2.8 fps

Grit Inf = 1.7 fps

Grit Eff = 1.93 fps

4/15/2013

59

Plant Grit ExperienceNTP Forced Vortex 24 ft dia

1203 Wet Metric Tons Removed from Digester 2008

NTP Forced Vortex 24 ft dia Removal Efficiency

Average hourly flow during sampling was approximately 20 MGD/hour

At 50 MGD, manufacturer specifies 95% removal of 100-mesh with 2.65 sg *

4/15/2013

60

System ConclusionsNansemond 24 ft Vortex

• Over 80% of grit was less than 297 µm

• System does not adequately protect down stream equipment.– 1203 Metric tons removed from digester in 2008

Overall ConclusionsTECHNIQUE- The grit sampling and testing methods are repeatable and meaningful- Evaluating the overall system is necessary

EQUIPMENT (for this study)- Forced vortex units had high efficiencies for large micron particles- Forced vortex units performed better at less than design flow rates

OPERATIONAL ADVICE- Evaluate putting more units in service to catch slug loads (first morning

flows and wet weather events)

FUTURE- More testing is desired during wet weather events- CFD Modeling to examine how to improve performance

4/15/2013

61

True Grit Summary

Questions?

Brian McNamaraArmy Base Treatment Plant

401 Lagoon RoadNorfolk, VA 23505

[email protected]

4/15/2013

62

Questions?

This Webcast Is Brought To You By:

Smith & Loveless Inc.

fundamentals of grit removal and grit system evaluation_part 2_color very good

Documents

grit size

grit properties

buoyant grit particle

distribution of grit

recent grit opinions

blackened grit particles

fundamentals of grit

clean water sand