unburned carbon in flyash measurement - scf ingenieria · unburned carbon in flyash measurement....

Stefan Kranz

Page: 1

we focus on your process

Unburned Carbon

In Flyash

Measurement

Stefan Kranz

Page: 2


Steam generator

BurnersS

econdary

air Air preheater

FD Fans

Combustion Air

Electrostatic

precipitator

Fly ash Silo

Fly ash

Flue gas

Coal bunker

Coal

mill

Pulverized

fuel

Prim

ary

air

MECONTROL

Coal

MECONTROL

Air

MECONTROL

UBC

System Application for Boiler Optimization

Stefan Kranz

Page: 3


Efficiency

Optimization of coal flow distribution, excess air and unburned carbon,

-------Reduction of flue gas losses and ignition losses-------

O2 Distribution

Equalization of the both O2 levels at the flue gas duct

Unburned Carbon (UBC) Reduction

Increase combustion efficiency, maximize fly ash quality

NOx

Reduction of NOx emission / ammonia use in DeNOx system (SCR)

CO

Prevention of boiler wall corrosion/slagging

Targets in Boiler Optimization

Stefan Kranz

Page: 4


Unburnt carbon in fly ash

Main indicator for combustion quality

Determination of further fly ash usage

Stefan Kranz

Page: 5


Fly ash can cost or bring benefits

Stefan Kranz

Page: 6


Internal Quality Assessment

Europe: According to EN-450 (DIN 18 990)

- Internal QA and coordination with customer

Quality has to be traceable at any stage of the fly ash

production. Sampling > 2 /day

Clearance of ``ready to ship`` ash according to DIN EN 450

3 Qualities

Category A < 5%

Category B > 2 % and < 7 %

Category C > 4% and < 9%

Sampling locations to be coordinated with the external auditor

(LOI, Particle Size, CaO)

Stefan Kranz

Page: 7


Internal Quality Assessment

North America: ASTM C 618

- Internal QA and coordination with customer

Quality has to be traceable at any stage of the fly ash

production. Sampling > 2 /day

Clearance of ``ready to ship`` ash according to DIN EN 450

Qualities

Category C < 6%

Sampling locations to be coordinated with the external auditor

(LOI, Particle Size, CaO)

Stefan Kranz

Page: 8


Analytik und Betriebsüberwachung ANA

Internal Quality control according to DIN 18 990

Online-C-Measurement

wet chemistry

(Calibration of Online measure

ment and accuracy check)

LOI

Particle Size

Free Lime

Chlorine; Sulfur Trioxide

Less frequent Checking:

Internal Self Auditing

(Self Auditing: has to comply

with formal procedure

of - DIN EN 450)

Frequent Checking:

Internal QA (this internal QA

needs to be trackable. It does

not need to comply with

formal procedures of EN 450)

Stefan Kranz

Page: 9



Proof of compliance

Source:Draft VGB-AK Werksqualitätshandbuch DIN 18990

Internal QA

Auditor

External

Auditor

Internal Quality Control

Power station

Sampling and QA procedures

Corrective Action

Assessment of QA supervision

tools

QA Historical Data

Internal Quality assessment

Producer

Process

control

Measurement

and accuracy

check

Storage,

Bulk handling

Checking of fly

ash quality

Take action in

case of

deviations

Checking of

equipment

accuracy

Supervision of

compliance

status

Stefan Kranz

Page: 10



Unburned Carbon / LOI

Internal QA of Fly Ash

0

1

2

3

4

5

6

0 1 2 3 4 5 6 7

LOI [wt-%]

UBC (TOC)

[wt%]

Stefan Kranz

Page: 11


Classification of Loss On Ignition measurement

LaboratorySamplers/ Instrumentation

Stefan Kranz

Page: 12


Classification of Loss On Ignition measurement

In-situ = no samplingSampling required

Stefan Kranz

Page: 13


Correlation between UBC, flue gas and excess air as a

function of the losses

UBC

exhaust gas

Losses

Flue gas

UBC

Optimum

Optimization Principle

Stefan Kranz

Page: 14


UBC Service

Focus/

ConditionsFlue gas losses

Optimization

Test

High Excess Air and

Low UBC

Reduction of the excess air (OFA or burner

air, SA)Base line testing

and optimization

Low Excess Air and

High UBC

Operating feedback on effectiveness of air

registersBase line testing

and optimization

High Excess Air and

High UBC

feedback on effectiveness of mills and

classifier settings and PA velocitiesBase line testing

and optimization

Optimization Optimization using DCS & OptimizersAdvice on control

scheme

modification

MECONTROL Boiler Optimization

Stefan Kranz

Page: 15


How to measure?

Stefan Kranz

Page: 16


Methods to estimate unburnt carbon in fly ash

Loss on ignition (LOI)

Laboratory Simple to conduct at laboratory without special analyzers. (Only furnace and scale required)

Most common laboratory analysis method for UBC estimation in power plants

Readings differ approximately 0,3 % to 1% wt UBC due to other reactands

Process Complicated to apply for online measurement (Sample extraction, piping, clogging problems)

Other reductants than carbon e.g. sulphur influence result

Stefan Kranz

Page: 17



TOC Laboratory:

Sample is evacuated and then incinerated with pure oxygen

CO2 concentration is measured by NDIR gas analyzer

very accurate (typically better than 0,1% wt UBC)

Expensive laboratory analyzer

Process:

Complicated to apply for online measurement (Sample

extraction, piping, clogging problems)

Maintenance & Cost intense

Limited reliability

Consumables, disposals

Stefan Kranz

Page: 18



Infra Red reflectance

Process: Sample is taken from process and exposed to infra red light

The reflectance is a measure for the unburnt carbon content

Complicated to apply for online measurement (Sample extraction, piping, clogging problems)

Dirt sensitive

Strong influence by particle size, raw coal type and other elements contained in the coal (Fe etc)

Stefan Kranz

Page: 19


Principle: Light Reflection

Typical size: 100µm

Reflection only in surface zone

Ash color (Fe2O3) has influence

Smaller carbon particles overrepresented!

Change in particle size spectrum between carbon and ash brings change in light reflection

free space surface zone

light quantum

Dark zone – no measurement

Stefan Kranz

Page: 20



0

0 200 400 600 800 1000

[%]

UB

C C

on

ten

t

[µm] Particle Size

Typical Correlation of UBC and Particle Size

The UBC particles are larger than the average of the ash particles

Stefan Kranz

Page: 21


same ash, same UBC content, different particle size

grinding ash especially makes carbon particles smaller

so the black particles get a larger surface

so the ash turns darker, even with the same UBC content

Stefan Kranz

Page: 22


Extractive Microwave System

Stefan Kranz

Page: 23


Microwave resonance/ Microwave absorption

Laboratory Microwave resonant frequency/ Microwave absorption depending on unburnt carbons

Process (extractive methods) Complicated to apply for online measurement (Sample extraction, piping, clogging problems)

Undefined bulk density

Maintenance & Cost intense

Extractive Microwave System

Stefan Kranz

Page: 24


In-Situ Microwave Resonance

solution

Stefan Kranz

Page: 25



In-Situ Microwave Resonance

Internationally patented In situ measurement

Fast accurate and reliable

Virtually no maintenance required

Fully self contained unit

Stefan Kranz

Page: 26


Measurement Principle

Dielectric constant of fly ash is a function of

the carbon content. Measuring the shift of

frequency in a resonator (f) the carbon

content can be calculated.

UBC = A + B f

A and B are the

calibration coefficients

Stefan Kranz

Page: 27


Principle: Microwave ResonanceFrequency fo,wavelength 1

Frequency fo,wavelength 2

no light quantums but

classical microwaves

penetration through all

material and inside the

particles

no impact of ash colour

no impact of particle size

low UBC

high UBC

Stefan Kranz

Page: 28


Oscillating circuit

Cεfres

1

HF-

Emission

C

HF-

Reception

Ash with carbon

ε >1

Ash with carbon

ε >1

Relation:

Resonance frequency depends on

capacity C and dielectricity ε

Air: ε = 1,

Ash with no carbon: e.g. ε = 2,25

Measurement Principle

Stefan Kranz

Page: 29


Measurement

Chamber

Auger

Sampling Tube

Sample will be compressed by a defined force in measurement

chamber

Large sample volume (500 ml)

Simple mechanical design

International patents

Sensor Arrangement

Stefan Kranz

Page: 30


Main Panel

Touch Panel controled

Protection up to IP66

Can handle up to 4 independent

UBC sensors

4-20mA I/Os to DCS

Profibus/Modbus/Ethernet capability

Remote maintenance access

Stefan Kranz

Page: 31


Measurement

Cabinet

Up to 8 SensorsPROMECON

Local control box Sensor(s)

System - Design

Stefan Kranz

Page: 32


Local control box

1: Power indicator

2: Error indicator

3: Release signal indicator

4: Filling indicator and button

5: Emptying indicator and button

6: Sampling indicator and button

7: Automatic/manual mode switch

8: Automatic mode indicator

9: Main switch

Power Error Release

Filling Emptying Sample

Auto / Hand

Main switch

Automatic

Stefan Kranz

Page: 33


In field applications

Stefan Kranz

Page: 34


Design UBC Sensor Type 300

Ash Auger

Sensor Flange Mounting

Measurement Chamber

Stefan Kranz

Page: 35


SCADA Screen for sampling location

Stefan Kranz

Page: 36


System Data Flow: Single Cabinet

Stefan Kranz

Page: 37


Multiple Cabinets

Stefan Kranz

Page: 38


A B

90% separation

90% separation

90% separation

90% separation 90% of the total fly ash

9,0%

0,9%

0,09%

Measurement Cabinet

4 Sensors

Sensor

Control Box

Sensor

Control Box

Sensor

Control Box

Sensor

Control Box

Sensor 1

Sensor 2

Sensor 3

Sensor 4

Digital Input Signals

Release signals per channel

(if necessary)

Promecon

AshFLO Sensors

4-20 mA Signals

(UBC)

0600 1200 1800

2400

3

4

5

6

UBC [%]

Microw. Generator

Microwave Receiver

PLC Unit

Power Supply

Signal Converter

Control Room

Touch Screen

Operator Interface

Fault Messages

Maximum Distance: 190m

Recommended measurement locations

Stefan Kranz

Page: 39


Measurement Location

Electric

precipitator

Fly ash

Flue gas

90% 9% 0.9%

MECONTROL UBC

Air Preheater

Spot sampling

100%

Weak Correlation to Silos

Strong Correlation to Silos

Silos

Flue gas

sampler

Not representative Representative

Stefan Kranz

Page: 40


sampler

Flue Gas Duct

Nozzle

High CO/UBC rope

Sampling

location

Ash and gas not homogenizedAsh discharged in

dense quantities

Cross sectional coverage:

0.000005%

Cross sectional coverage: 2-8%

Fly ash concentration: 5g/m3

Fly ash concentration:

200,000 g/m3

Sampling of the Ash Flow

Ash Hopper

Stefan Kranz

Page: 41


Fly Ash Collection Rate

ESP Hopper

1g/minute 100 g/minute

An adequate response time in the flue

gas duct is only achievable by a very

small sample size

Sampling can achieve both:

Large, representative

sample size

fast response measurement

ash travel time from burner

to DUCT sampler: 3 secs

ash collection time: 5

minutes

ash travel time from burner to

PROMECON sampler: 15 secs

ash collection time: 5 minutes (hopper cycle)

Sampling synchronous to cycle

Measurement Response Time

Flue Gas Duct

Stefan Kranz

Page: 42


Sample Size

5 g out of flue gas duct 500 g out of precipitator

Stefan Kranz

Page: 43


Extractive MeasurementIn-Situ Measurement

160oC

Flue gas

duct

Measurement Panel

Ambient Temperature

cyclone

separation

classification?

Transport

through

heated

pipes

MW measure-

ment of Air/

Ash mixture

Static weighing of 5g sample

60oC

Sampling and measurement

in the hopper:

no Extraction,

no cyclone separation

no static weighing

no heating elements

ESP

Hopper

Stefan Kranz

Page: 44


All-in-one sensor

non extractive sampling!

no pneumatic transport

1 moving part

simple/robust design

no weighing

Mechanical sampling system

Mechanical: Less moving parts are better

Stefan Kranz

Page: 45


Unequal particle size distribution

Bigger particles drop faster than small particles

Unqual particle size distribution in horizontal conveyor

pipes

Unequal UBC content depending on particle size

Inaccurate measurement since sample is not

representative

Stefan Kranz

Page: 46


Measurement Location

Electric

precipitator

Fly ash

Flue gas

90% 9% 0.9%

MECONTROL UBC

Air Preheater

100%

Silos

Not representativeRepresentative

Sampling from

conveyor pipe

Stefan Kranz

Page: 47


Conclusion measurement location

No particle size classification/ fractionating

Representative averaged sample

Short response time

Stefan Kranz

Page: 48


Basic

Accuracy comparison with

Laboratory TOC Analyzer

Stefan Kranz

Page: 49


0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8 10

Block1 Sensor1

Block1 Sensor2

Block2 Sensor1

Block2 Sensor2

-0,60%

+0,60%

Unit 1 Sensor 1

Unit 1 Sensor 2

Unit 2 Sensor 1

Unit 2 Sensor 2

Typical Measurement Data

UBC (lab)

UBC

Promecon

Stefan Kranz

Page: 50


Stefan Kranz

Page: 51


Measurement Data of

0

100

200

300

400

500

600

700

800

03.07.2001

06:00

08.07.2001

06:00

13.07.2001

06:00

18.07.2001

06:00

23.07.2001

06:00

28.07.2001

06:00

02.08.2001

06:00

07.08.2001

06:00

12.08.2001

06:00

Kanal 0 Labor 0

0

100

200

300

400

500

600

700

800

900

03.07.2001 06:00 03.07.2001 12:00 03.07.2001 18:00 04.07.2001 00:00 04.07.2001 06:00 04.07.2001 12:00

Kanal 2 Labor 2

Stefan Kranz

Page: 52


Wedel

0

100

200

300

400

500

600

700

800

900

1000

18.07.2001

00:00

20.07.2001

00:00

22.07.2001

00:00

24.07.2001

00:00

26.07.2001

00:00

28.07.2001

00:00

Re

st-

C [

0,0

1%

]

0

100

200

300

400

500

600

700

800

900

1000

Kanal 2

Kanal 3

Temperatur Kanal 2

Temperatur Kanal 3

Measurement Data of

Stefan Kranz

Page: 53


Trial Run at "Reuter West" Plant, dated 2001-11-12

-170

-165

-160

-155

-150

-145

-140

-135

-130

-125

06:00 07:00 08:00 09:00 10:00 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00

Time

Se

co

nd

ary

Air

x 1

00

0 (

inve

rte

d)

[m³/

h]

ST

P

1

2

3

4

5

6

7

8

9

10

UB

C [

%]

Secondary Air (MECONTROL Air)

UBC ETG 17 (MECONTROL UBC)

UBC ETG 18 (MECONTROL UBC)

Remark: Listed values of secondary air

amount are only for one burner plane.

1 % O2

Boiler/Mill Optimization w/ UBC Monitoring

Stefan Kranz

Page: 54


Optimization Results

Trial run at "Reuter West” power plant, dated 2001-11-12

2

4

6

8

10

12

14

16

18

UB

C [

%]

07:00 7:30 8:00 8:30 9:00 9:30 10:00 10:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00 15:30 16:00 16:30 17:00

Time

0

0,5

1

1.5

2

2.5

3

3.5

4

4.5

O2

[%]

UBC Fly ash (MECONTROL UBC)

UBC Fly ash (Lab analysis)

O2 Boiler / DeNOx outlet

Excess Air Reduction

Stefan Kranz

Page: 55


Excess Air Reduction Jan. 22, 2002

2

2,5

3

3,5

4

4,5

5

10:00 11:12 12:24 13:36 14:48 16:00 17:12

240

245

250

255

260

265

270

275

280

seco

nd

ary

air

[kg

/s]

O2 Boiler out SA total

SA2 TA SA1+CA

CA

SA1

SA2

TA

TA

6 Burners x 4 levels

O 2

[Vo

l.-%

]

Power Station Farge, Power Utility E-on

Stefan Kranz

Page: 56


2

2,5

3

3,5

4

4,5

5

10:00 11:12 12:24 13:36 14:48 16:00 17:12

O 2

[Vo

l.-%

]

500

520

540

560

580

600

620

NO

x [

mg

/m³]

O2 boiler out NOx before SCR

SA2 TA SA1+CA

Power Station Farge, Power Utility E-on

Stefan Kranz

Page: 57


ρ

Integrated density correction by nucleonic source

Enhanced online accuracy

Independent of ash quality variation

Easy upgrade of existing systems

Stefan Kranz

Page: 58


ρ

Sensor arrangement

Stefan Kranz

Page: 59


ρ

Don‘t fear about radioactivity:

Very small source 185 MBq

Savely sealed, Stainless steel shell, lead & steel

protector

Typically used in Power stations for

Density measurement of lime milk (FGD)

Level switch for coal bunker exit

Belt scale measurements for feeder

Nucleonic Source

Stefan Kranz

Page: 60


Typical accuracy

Laboratory - UBC correlation (density corrected)

0

1

2

3

4

5

6

7

8

9

10

11

12

0 1 2 3 4 5 6 7 8 9 10 11 12

TOC

UB

C Ash Samples

Regression

Sample

No.

TOC

Lab

UBC

reading Accuracy

1 10,49 10,52 -0,03

2 1,65 1,89 -0,24

3 1,80 1,53 0,27

4 1,96 1,90 0,05

5 1,35 1,53 -0,18

6 4,44 4,31 0,13

Accuracy process versus laboratory < 0,2% wt

ρ

Stefan Kranz

Page: 61


Real world examples

without ρ correction

σ = 1,10%

No. Labor UBC Accuracy

1 10,49 7,12 3,37

2 1,65 2,36 -0,71

3 1,80 2,53 -0,73

4 1,96 2,69 -0,73

5 1,35 0,91 0,43

6 4,44 4,37 0,07

7 4,97 4,13 0,84

8 5,00 4,16 0,84

9 5,04 4,12 0,92

10 4,28 4,15 0,13

11 4,36 4,17 0,19

12 3,16 2,36 0,80

13 3,00 2,10 0,90

14 3,05 2,27 0,78

15 2,85 2,31 0,54

16 3,40 4,53 -1,13

17 3,30 4,50 -1,20

18 2,70 3,10 -0,40

Laboratory - UBC correlation

Without density correction

0

1

2

3

4

5

6

7

8

9

10

11

12

0 1 2 3 4 5 6 7 8 9 10 11 12

Laboratory

UB

C

Regression

Shikoku

ResearchYonghung 1

Yonghung 2

Yonghung 3

Scholven C

Stefan Kranz

Page: 62


ρ

Accuracy process versus laboratory ~ 0,2%

No. Lab UBC Accuracy

1 10,49 10,48 0,01

2 1,65 1,88 -0,23

3 1,80 1,53 0,27

4 1,96 1,90 0,06

5 1,35 1,52 -0,18

6 4,44 4,29 0,14

7 4,97 4,91 0,06

8 5,00 5,07 -0,07

9 5,04 5,10 -0,06

10 4,28 4,17 0,11

11 4,36 4,22 0,14

12 3,16 2,90 0,26

13 3,00 2,82 0,18

14 3,05 2,97 0,08

15 2,85 2,79 0,06

16 3,40 3,81 -0,41

17 3,30 3,47 -0,17

18 2,70 2,96 -0,26

σ = 0,19%

Laboratory - UBC correlation

Density corrected

0

1

2

3

4

5

6

7

8

9

10

11

12

0 1 2 3 4 5 6 7 8 9 10 11 12

Laboratory

UB

C

Regression

Shikoku

ResearchYonghung 1

Yonghung 2

Yonghung 3

Scholven C

Real world examples

with ρ correction

Stefan Kranz

Page: 63


Accuracy

Laboratory

0,1%

Gain of the effective accuracy

2 samples/ day

Accuracy

MECONTROL UBC

0,6%

50 samples/ day

(situation Altbach)

n

σσ

SystemMeanValueExpected

%07,02

%1,0MeanValueExpectedσ %08,0

50

%6,0MeanValueExpectedσ

Gain of effective accuracy due

to frequent measurements

Stefan Kranz

Page: 64


Comparison Laboratory vs. Online

Altbach HKW2

0

1

2

3

4

5

8.10 9.10

UB

C (

%)

UBC

Channel 1

Lab

Stefan Kranz

Page: 65


On-line in situ measurement of UBC

Fully self contained unit, no extractive sampling,

simple mechanical design

High accuracy and absolute measurement of UBC

Virtually no maintenance, low LTCO

Independent of ash density

Large international installed basis > 200 Sensors at >

100 boilers

Enhanced accuracy version 0,2% online vs. lab

Easy upgrade of existing installations

Conclusions

ρ

Stefan Kranz

Page: 66


Accurate measurement of the key combustion

parameter

Optimization of mill &boiler performance

Improvement of NOx, CO, O2 and UBC

High cost savings by reduction of primary losses

Cost efficient fly ash disposal

Recommendations by Babcock Borsig Power

(BBP) and current utility users

Benefits to the Power Plant

Stefan Kranz

Page: 67


PROMECON is the world leader with MECONTROL UBC

installations in more than 90 Power Plants worldwide.

At several international utilities PROMECON has achieved

improvements of several $100,000 US per year by its systems.

Recent investigation at reference Power Stations

Optimization of the boiler efficiency: 0.4 – 1.0 %

Reduction of the unburned carbon: 10 - 30 %

Reduction of CO: 50 - 80 %

Saving coal consumption: $300,000 /y

PROMECON estimates $150,000 – $500,000 US with NOx optimization and

boiler wear reduction.

Benefits with PROMECON systems

Stefan Kranz

Page: 68


Thank you very much!

Stefan Kranz

Page: 69


unburned carbon in flyash measurement - scf ingenieria · unburned carbon in flyash measurement....

Documents