0

0.5

1

1.5

2

2.5

3

Study Objective & Intro• Resolve Problem of Effluent Total

Phosphorus SPIKING in a Biological Phosphorus Removal (BPR) System

• Proposed Solution – Automatically Feeding a high rbCOD product

• LaCrosse - Logical Place for Study– Has Effluent TP Spiking -- Already Using

high rbCOD product on Weekends– Close to Home

La Crosse WWTP

• Design - 20 MGD• Present - 10 MGD • Liquid - Fine Screen/Grit/Primaries/BPR/UV • Solids – Co-Thicken Gravity

Thickeners/Anaerobic Digestion/GBT Thickening/Storage/Liquid Injection

• Major Contributor – City Brewery, high rbCOD

PRI ANAEROBIC ZONE

ANOXIC ZONE

AEROBIC ZONE

EFF

Molasses FormulateProductHigh rbCOD Product

added to Primary Clarifier Effluent Trough

0

0.5

1

1.5

2

2.5

3

6/2/

2011

Th

u

6/3/

2011

Fr

i

6/4/

2011

Sa

t

6/5/

2011

Su

n

6/6/

2011

M

on

6/7/

2011

Tu

e

6/8/

2011

W

ed

6/9/

2011

Th

u

6/10

/201

1 F

ri

6/11

/201

1 S

at

6/12

/201

1 S

un

6/13

/201

1 M

on

6/14

/201

1 T

ue

6/15

/201

1 W

ed

6/16

/201

1 T

hu

6/17

/201

1 F

ri

6/18

/201

1 S

at

6/19

/201

1 S

un

6/20

/201

1 M

on

6/21

/201

1 T

ue

6/22

/201

1 W

ed

6/23

/201

1 T

hu

6/24

/201

1 F

ri

6/25

/201

1 S

at

6/26

/201

1 S

un

6/27

/201

1 M

on

6/28

/201

1 T

ue

6/29

/201

1 W

ed

6/30

/201

1 T

hu

7/1/

2011

Fr

i

7/2/

2011

Sa

t

7/3/

2011

Su

n

7/4/

2011

M

on

7/5/

2011

Tu

e

7/6/

2011

W

ed

BPR Works Well 75% of

Time

Eff Spiking P on Weekends or Beginning of

Week

BPR Operational Difficulties

Investigate/Analyze Difficulties

• Survey/Investigate BPR systems using P profile testing–What is P profile testing?

• Snapshot in time–Why use it for this investigation

• P-Release in Anaerobic Zone • P-Uptake in Aerobic Zone

–Are there Trends?• Simple analyses– on-line analyzers available

P-Release in Anaerobic Zone WEF (2010), Nutrient Removal WEF Manual of

Practice No. 34 (Page 523, Section 2.3.4)

• “..higher overall release..better Premoval”

• "The release of P can result in Pconcentration in the anaerobic zonefrom the influent concentration and still result in good Puptake in the aeration basin."

Performing P Profiles - BPR Systems

• Ran Multiple P Profiles on LaCrosse System and at numerous other WWTPs

• Taking pulse of system • Multiple profiles/day multiple days -

weekends

QLF Specialty Products…Phosphorus Profiles of a BPR System Which has Lost and Then Regained rbCOD Loading Over a Weekend

Analyze Results for Patterns• Need a Mathematical Expression

–For a baseline to measurement BPR performance

• Back to ANAEROBIC ZONE - Healthy BPR System–PAO’s uptake VFAs and P Release–Small P-Release – low influent rbCOD–High P-Release – high influent rbCOD

Analyze Results for Patterns

• Using ONLY anaerobic zone P release as a sole measure of BPR health had problems;–Changing influent P concentrations

•Side streams can give False High P-Release

–Need to Compare anaerobic P-Release to something--WEF

Analyze Results for Patterns

• Created Mathematical Expression • Anaerobic Zone P-Release divided by

BPR Influent P • For this study we will refer to it as

Phosphorus Release Ratio - PR2

• Can PR2 be used to measure the health and performance of a BPR system? WEF - PR2 between 2 and 4 GOOD

0.00

0.50

1.00

1.50

2.00

2.50

Mon

9:1

5 AM

Mon

11:

20 A

M

Mon

1:1

8 PM

Mon

3:0

0 PM

Mon

4:1

5 PM

Mon

7:3

0 PM

Mon

11:

00 P

M

Tues

3:5

0 AM

Tues

7:2

5 AM

Tues

9:5

0 AM

Tues

1:2

0 PM

PO4

as P

, mg/

l and

PR2

P Uptake Compared to PR2 - Is NOT Adjusted for Detention Time

P Release Ratio, PR2 Aerobic Zone 3 OP as P, mg/l

These 2 lines appear to correlate

well, but base on the what's going on

THEY SHOULDN'T

Anaerobic Zone P divided by Influent P = P-Release Ratio (PR2)

Measurement of Satisfying BPR rbCOD Demand

Last Aerobic ZonePO4 as P, mg/l

Measurement of BPR Performance

0.00

0.50

1.00

1.50

2.00

2.50

Mon

7:1

0 AM

Mon

9:1

5 AM

Mon

11:

20 A

M

Mon

1:1

8 PM

Mon

3:0

0 PM

Mon

4:1

5 PM

Mon

7:3

0 PM

Mon

11:

00 P

M

Tues

3:5

0 AM

Tues

7:2

5 AM

PO4

as P

, mg/

l and

PR2

P Uptake Compared to PR2 - Is Adjusted for Detention Time

P Release Ratio, PR2 Aerobic Zone 3 OP as P, mg/l

Anaerobic Zone P divided by Influent P = P-Release Ratio (PR2)

Measurement of Satisfying BPR rbCOD Demand

Last Aerobic ZonePO4 as P, mg/l

Measurement of BPR Performance

Multiple P Profiles Adjusted for System Detention Time were

Performed - Results

• Increase/Decrease of rbCOD has almost instantaneous effect on Anaerobic Zone P-Release - PR2

• PR2 Appears to Predict BPR Performance - Effluent DRP Concentrations

Plans and Setup for Trial

• Related PR2 findings to Method of Automatic Control

• Added Ramp Control - SCADA system • Needed: An on-line analyzer – monitor

2 sample points with one analyzer• Worked with Instrument vendor to

supply an analyzer

Primary Effluent orthophosphate

Input, mg/l

Anaerobic Zoneorthophosphate

Input, mg/l

PLC Controller Calculate

P-Release Ratio

Aeration Basin Effluent

orthophosphate Input, mg/l

Output Control EnhanceBioP+N - Feed

Based on P-Release Ratio Set Point

Ferric - PID Loop ControlMolasses Formulation

Feed Pump to Primary EffluentFerric Chloride Feed Pump to Aeration Tank Effluent Tough

On-Line Analyzer Info

2 sample points In Situ filters Analyzes for orthophosphorus and nitrate Sampling Frequency – one every 15 minutes Analyte data can be used for monitoring

and control

Aerobic Zone 2 Aerobic Zone 3

Anaerobic Zone 1

Anoxic ZoneAerobic Zone 1

Anaerobic Zone 2Inf

RAS

2

1

3

4

1 - BPR Influent Analyzer Sample Point2 - Anaerobic Zone Analyzer Sample Point3 - Nitrate Recycle Pump4 - There is not wall at this location. It is divided into 2 zones for sampling purposes

La Crosse WWTP BPR System A2/O

12

31. BPR influent sample point, primary effluent was pumped through drum to provide this sample point2. Anaerobic Zone sampling point3. Anaerobic Zone sample filter

Trial• Length – 4 weeks

– Focus on the weekends when rbCOD loading changes–

– DID FEED AUTOMATICALLY DURING WEEK TOO AT HIGHER PR2 SET POINT

• First Worked out Bugs, took 3 weeks– Learning the best dose setting required

trial/error– Accidently let on-line analyzer reagent run out– Etc

0

1

2

3

4

5

6

7

8

9

10

0

5

10

15

20

25

30

35

40

45

50

Thu

Mid

nigh

tFr

i 2Fr

i 4Fr

i 6Fr

i 8Fr

i 10

Fri N

oon

Fri 1

4Fr

i 16

Fri 1

8Fr

i 20

Fri 2

2Fr

i Mid

nigh

tSa

t 2Sa

t4Sa

t6Sa

t8Sa

t10

Sat N

oon

Sat1

4Sa

t16

Sat1

8Sa

t20

Sat2

2Sa

t Mid

nigh

tSu

n 2

Sun

4Su

n 6

Sun

8Su

n 10

Sun

Noo

nSu

n 14

Sun

16Su

n 18

Sun

20Su

n 22

Sun

Mid

nigh

tM

on 2

Mon

4M

on 6

Mon

8M

on 1

0M

on N

oon

Mon

14

PR2

Ratio

-O

rtho

-P, m

g/l

rbCO

D Pr

oduc

t Pum

p Sp

eed,

%

rbCOD Product Pump Speed = % PR2-Ratio Eff Online Ortho-P Avg PR2 Ratio 2 per. Mov. Avg. (PR2-Ratio)

1-PR2 set point too low and ramp control set too react slow.2-PR2, QLF Pump Speed, and Effluent OP - level out3-rbCOD Product - OFF – PR2 ↓ Effluent OP ↑ 4-Monday - Industrial rbCOD loading returns, drives PR2 ↑ and Effluent OP ↓

1

2

3

4

0

1

2

3

4

5

6

7

8

9

10

0

5

10

15

20

25

30

35

40

45

50

Thu

Noo

nTh

u 15

Thu

18Th

u 21

Thu

Mid

nigh

tFr

i 3Fr

i 6Fr

i 9Fr

i Noo

nFr

i 15

Fri 1

8Fr

i 21

Fri M

idni

ght

Sat3

Sat6

Sat9

Sat N

oon

Sat1

5Sa

t18

Sat2

1Sa

t Mid

nigh

tSu

n 3

Sun

6Su

n 9

Sun

Noo

nSu

n 15

Sun

18Su

n 21

Sun

Mid

nigh

tM

on 3

Mon

6M

on 9

Mon

Noo

nM

on 1

5M

on 1

8M

on 2

1M

on M

idni

ght

Tue

3Tu

e 6

Tue

9Tu

e N

oon

Tue

15Tu

e 18

PR2

Ratio

-O

rtho

-P, m

g/l

rbCO

D Pr

oduc

t Pum

p Sp

eed,

%

rbCOD Product Pump Speed = % PR2-Ratio Final Eff. Ortho P Avg PR2-Ratio 2 per. Mov. Avg. (PR2-Ratio)

All Control Settings Fine Tuned Successful Run

Conclusion• Successful - Fed high rbCOD product

automatically to maintain consistently low effluent DRP

• Each BPR system will have its own optimum PR2 and its own unique rbCOD Demand to achieve that PR2

• Auto feed high rbCOD system will not FIX Extreme rbCOD shock loads, operational issues and toxic loads - cover with automatic chemical feed system

Conclusion• Benefits related to automated feed

– Keep biology HAPPY – THEN IT PERFORMS– Providing rbCOD when necessary– Process covers multiple BPR sins

• Nitrates in RAS or from recycle ammonia spikes – DATA FROM ANALYZER

• Oxygen in RAS• P spikes in recycle

– System works on rbCOD DEMAND not on influent rbCOD:TP ratio• Stops feeding – When Influent rbCOD is

Adequate or High

Other Potential Benefits of Automatically Feeding High rbCOD

These need more investigation/assessment• Reduced sludge production when

compared to using chemical for P control• Reduced effluent TSS because biological

loading to system is stabilized–Eff TSS during the successful weekend of

auto feeding study was very low – 1 to 2 mg/l usually 3 to 5 mg/l - FIRST TIME EVER

Other Potential Benefits of Automatically Feeding High rbCOD

These need more investigation/assessment• Reduced heavy metals in sludge due to

reduced ferric requirement • Improved solids dewatering when

compared to systems that use alum

Closing Thoughts • How much rbCOD does a system need, and when?• BPR rbCOD demand (PR2) vs. rbCOD:TP• Compare Chlorine Residual and PR2

– Chlorine residual is an indirect operational test(Fecal) is demand met? -- nitrite ↑ demand

– PR2 may also be an indirect measure of how much the BPR rbCOD Demand is satisfied• BPR systems have Real time Demands based on

changing conditions in anaerobic zone• Changing rbCOD, P, NO3 and O2 loadings

• Establish the optimal PR2 by comparing the PR2 levels to effluent DRP

Jared Greeno, greenoja@cityoflacrosse.orgGreg Paul, greg@QLF.comDavid Roskowic, David@chemscan.com