ON-LINE MEASUREMENT OF PULVERIZED COAL

By
Dave Earley
Air Monitor Corporation
Dave Earley
Air Monitor Corporation
178 Mine Lake Court
Raleigh, North Carolina 27615
Abstract
Coal-fired electric utilities consistently struggle with attempts to improve overall plant performance by
achieving optimum combustion. While many techniques are employed, little has been done to optimize
combustion at the individual burners. Distribution of windbox airflow and pulverized coal flow can vary
greatly. There has been no effective method to measure coal and air, and the utility industry continues to
accept these performance inadequacies.
In the age of deregulation and with increasing concerns over emissions, the utility industry continues to
search for better methods of fuel and airflow measurement and control. This is especially true with the
use of low NO
x
burners, which require accurate airflow and fuel balance for optimum reduction of NO
x
while simultaneously minimizing unburned carbon.
In 1997, a large Utility in Germany tested the use of a new coal flow measuring device which utilizes low
frequency microwaves to accurately measure the absolute mass flow in coal pipes. When applied to
coal outlets from a pulverizer, this device can accurately measure coal flow distribution from pipe-to-
pipe. This device has successfully proven its ability to measure coal flow distribution with no
maintenance drift problems. Based on the devices success on one mill, the Utility elected to equip all of
the pipes on one boiler at this station. Secondary air (SA) is individually ducted to each burner on this
boiler (unlike SA in the United States); the plant will control airflow to account for fuel imbalances on-
line in an attempt to increase plant efficiency by reducing excess oxygen.
Introduction
In the United States, the Clean Air Act and deregulation prove to be driving forces in electric utility plant
improvements. One of the more important factors that addresses both issues is fuel/air ratio, which,
prior to the Clean Air Act and deregulation, was rarely scrutinized. Not only is the fuel/air ratio
important to optimizing combustion efficiency, it is also critical to the performance of low NO
x
burners
and combustion modifications associated with low NO
x
programs (e.g., staged combustion). Poor
fuel/air ratio can also lead to problems such as slagging, and high unburned carbon in flyash which can
impact ash sales.
Currently, the method for measuring coal flow into a boiler consists of volumetric and gravimetric coal
feeders that measure the coal entering each coal pulverizer. From the pulverizer, the coal is delivered to
multiple coal pipes, then, via the coal pipes; the pulverized coal is delivered to the individual burners.
Balancing of the coal flow distribution between burners has traditionally been attempted through clean or
dirty air traversing and the installation of orifice plates. These types of manual traverses can only
provide information at a single operating condition. Since it is widely accepted that the pipe-to-pipe
coal distribution changes with load and time, manual traverses, even if accurate, can not be sufficient for
balancing coal pipes to achieve optimum combustion.
While NO
x
improvements are not as critical to the German utilities, which typically have SCRs in place
to reduce NO
x
, a deregulated environment is forcing that industry to look at efficiency upgrades. One
large German Utility, which had previously determined via extractive sampling that their pipe-to-pipe
coal imbalance was severe, decided to use a new technology for measuring pulverized fuel flow in an
effort to reduce the overall excess air and thus improve efficiency. The new technology that was
selected and proven to work was Pf- FLO by Promecon.
Review of Pf-FLO Technology
Measurement Principle
Method. To obtain the mass flow of coal to a burner, one needs to know the concentration
1
and the
velocity of the pulverized coal in the burner pipe. Both of these quantities are measured independently
by Pf-FLO.
Concentration. The concentration of the pulverized fuel (Pf) is measured using the burner pipe as a
waveguide. The waveguide properties used are only dependent on the dielectric load, i.e. the density of
the Pf in the measuring section.
The use of very low resonance frequency microwaves has two advantages:
There is no possibility of reflected microwaves returning from outside the measuring section. It is a
common problem of microwave measurements that microwaves with high frequencies travel through
the pipework without loss and therefore can be reflected at far away points. The reflected wave is
usually a disturbance for microwave measurements.
The absorption of microwaves by water vapor increases exponentially with increasing frequencies.
2
These attenuation effects are negligible at the low frequencies of the Pf-FLO system.

1
The term 'concentration' is meant as mass concentration or mass density in this report.
2
It is therefore quite common to use microwave systems to measure free moisture in coal on a conveyer.
Velocity. The velocity of the pulverized coal is measured by a cross correlation method, which is
conceptually depicted in Figure I (see Figures following the References Section). Two sensors are
mounted at a known distance along a pipe. From each of the sensors a series of stochastic signals is
acquired, which have irregularities that relate to small coal density variations in the pipe. The signals to
both sensors are nearly identical but shifted by the time the Pf needs to get from one sensor to the other.
One can calculate this time shift with cross correlation. As the distance between the sensors is fixed, the
velocity of the pulverized fuel can be accurately calculated in the pipe cross section. The obtained
velocity is the velocity of the coal content in the pipe.
The Concept of Absolute Measurement
Pf-FLO measures the coal flow through each burner pipe absolutely, i.e. independent and
individual. This absolute measurement is a fundamental concept that has several advantages to the
user.
I ndependency. Each pipe is measured individually. The measurement values of one pipe will not be
influenced by the rest of the pipes. Similarly, a failure of the measurement in one pipe does not affect
the other pipes, and their obtained values are still valid.
Verification. The absolute measurement enables the user to verify the calibration of the system. One
can cross check the sum of all absolute measurement values of one mill with that mills feed rate. If an
increase in value of one channel is offset by the other channels of that mill and the sum continues to track
the feed rate, it provides confirmation of calibration stability. If the sum does not track the feed rate, it is
a strong sign that something with the system is wrong. This enables the user to detect any malfunction at
an early stage. Examples for both characteristics are given below.
Installation and Set-Up
Technology Principle
The hardware of the Pf-FLO system has two main elements: Microwave sensors and a Measurement
Cabinet.
Microwave Sensors
The measurement is performed by ''drill and tap'' sensors. Figure II (see Figures following the
References Section) shows a typical configuration of the sensors on a burner pipe.
Referencing Figure I I (see Figures following the References Section), the coal concentration
measurement is performed by transmitter and receiver pairs located along the pipe. Since the
polarization of the microwave signal from a transmitter/receiver pair is axially symmetrical yet has a blind
spot 90 to the orientation to the pair, the installation of a second transmitter/receiver pair located 90 to
the other allows the measurement of the entire pipe cross section. With the sensor pairs located in the
vertical and horizontal locations, it is also possible to detect the presence of coal layout in a horizontal
section of pipe. The fifth sensor (Rys) shown in Figure I I (see Figures following the References
Section) is used in conjunction with transmitter Ty and receiver Ry for velocity measurement.
The sensors are extremely abrasion resistant, proven by long-term coal pipe installations that have
produced no measurable signs of wear. Changing sensors is simply accomplished by unscrewing the
old sensor and installing a replacement, without need for any subsequent re-calibration. Figure I I I
(see Figures following the References Section) shows the picture of one sensor installed in a coal
pipe.
Measurement Cabinet
The Measurement Cabinet contains the microwave generator and receiver. The microwave signal is
multiplexed to the individual pipe sensor pairs, and in its base configuration the Measurement Cabinet
can monitor up to 8 pipes. Via a cabinet extension, the quantity of monitored pipes can be extended to
12. Multiple Measurement Cabinets can be networked together to allow monitoring of up to 144 pipes
at a central location. Each Measurement Cabinet operates on a stand-alone basis, utilizing a PLC to
control the sequence of measurements and processes all received inputs.
Maintenance
The system cyclically tests its own performance. If it detects any irregularities, an output warning signal
is generated for local monitoring, or remotely via the installed modem. Due
to the extreme stability of the microwave signal generator, there is no need for periodic recalibration, nor
is any scheduled maintenance required.
Health and Safety
All active parts of Pf-FLO have full CE approval. Pf-FLO does not emit any electromagnetic
radiation that could disturb any other instrument that itself is in compliance with the relevant EMC
(electromagnetic compatibility) guidelines. The maximum level of energy radiation emitted by the
equipment is significantly less than the average mobile telephone or a typical PC. The sensors are
specially grounded to prevent electrical energy transmission.
Data Analysis
In November 1997, Pf- FLO was installed on eight coal pipes from an MPS pulverizer at the
referenced German Utility. The unit was started up and the mass flows through the pipes were
evaluated (Figure I V - see Figures following the References Section). As suspected by the plant,
the pipe-to-pipe distribution was terrible. Because the data for each pipe is absolute (not dependent on
data from other coal pipes, mill feeder input or other outside source), there was a simple check to
determine that this data, and associated pipe maldistribution, was in fact correct the mass flow through
each pipe was summed and then compared with the coal feeder (Figure V - see Figures following
the References Section). Because the mass flow through the pipes is generated in dimensionless units,
the sum is also dimensionless. The sum is plotted on a graph showing tons/hour of coal as measured by
the feeder, and is thus shown with an arbitrary scaling. Additional on-line data can be seen in Figure
VI (see Figures following the References Section) and the corresponding sum of the pipes on
Figure VI I (see Figures following the References Section). The sum of the pipes continued to track
the feeder rate with the same offset. This comparison was done for over a year with similar results,
giving plant personnel the confidence in Pf-FLOs ability to measure absolute coal flow in all of their
coal pipes.
Correction for Problems
Data analysis provided an explanation of why some pipes were experiencing pluggage and some
burners had trouble staying ignited (blow-out). It also helped explain the tremendous excess oxygen
stratification that was being measured on the back-end of the boiler.
The results of the extensive product evaluation prompted the plant to purchase Pf-FLO for all coal
pipes on this boiler. Since secondary air is individually ducted to each burner at this plant, with
individual flow measurements and dampers available, the plant plans to first correct the gross pipe-to-
pipe coal flow imbalances, then perform on-line adjustment of coal flow imbalances associated with
load change and other factors by dynamically adjusting secondary airflow to the individual burners.
Other Installations
Since Pf-FLO was first implemented at this plant, several other European utilities have implemented
Pf-FLO with similar results.
In addition, there are currently nine Pf-FLO installations in progress within the United States. Since
individually ducted burners are not typically found in the United States, Air Monitor Corporation has
developed airflow measuring devices designed specifically for Individual Burner Airflow Measurement
(IBAM Patent Pending Insertion Probes). The IBAM probes are currently being used by
many utilities (over 2000 burner installations) to balance airflow between individual burners. When
combined with Pf-FLO, users will have the means with which to measure, and ultimately balance,
both air and fuel to individual burners.
In addition to measuring coal pipe imbalance, Pf-FLO has been used to detect coal layout and fuel
slugging, and measure coal velocity imbalance, all common problems associated with pulverized fuel
transport that negatively impact burner combustion performance. Field trials are currently underway in
Europe to evaluate Pf-FLOs capability to detect changes in particle fineness.
Current installation are proving that using Pf-FLO alone or together with IBAM will help improve
many plant concerns. Some examples of this are:
- Reduction of O
2
stratification from 7% to less than 1%, thus allowing for overall excess O
2
reduction.
- Reduction of NO
x
emissions by 10% or more.
- Reduction of CO from 400 ppm to less than 25 ppm.
- Reduction of LOI from 4% to less than 3% through coal flow and velocity adjustment.
- Reduce or eliminate occurrence of burner pluggage due to low coal velocity.
References
1. Earley, D., Penterson, C., Accurate Burner Airflow Measurement for Low NO
x
Burners
Presented at 1998 ISA POWID/EPRI Controls and Instrumentation Conference, Scottsdale, AZ.
2. Vierstra, S., Earley, D., Balancing Low NO
x
Burner Air Flows Through the Use of Individual
Burner Airflow Monitors. Presented at Power-Gen International 1998 Conference, Orlando, FL
correlation
time of flight
T
PF velocity
sensor
dist ance
signal 2
y(t)=x(t-T)
signal 1
x(t)
sensor
Figure I
Principle of the Velocity Measurement
Ty = Transmitter Y axis
Tx = Transmitter X axis
Ry = Receiver Y axis
Rx = Receiver X axis
Rys = 2
nd
Receiver Y axis
D = Pipe Diameter
Figure II
Typical Installation on a Burner Pipe
Figure III
Sensor on a Burner Pipe
Mass Flow through Pipes 1 to 8
0
200
400
600
800
1000
1200
1400
1600
1800
2
2
:
2
5
2
3
:
3
5
0
:
4
4
1
:
5
3
3
:
0
3
4
:
1
2
5
:
2
1
6
:
3
1
7
:
4
0
8
:
4
9
9
:
5
9
1
1
:
0
8
1
2
:
1
7
1
3
:
2
7
1
4
:
3
6
1
5
:
4
5
1
6
:
5
4
1
8
:
0
4
1
9
:
0
1
1
9
:
1
5
1
9
:
2
5
1
9
:
3
4
1
9
:
4
5
1
9
:
5
1
1
9
:
5
6
2
0
:
0
1
2
0
:
0
8
2
0
:
1
4
2
0
:
1
9
2
0
:
2
4
2
0
:
2
9
2
0
:
3
3
2
0
:
3
8
Time
Pipe 1 Pipe 2 Pipe 3 Pipe 4 Pipe 5 Pipe 6 Pipe 7 Pipe 8
Figure IV
Figure V
Figure VI
Feed Rate / Sum of All Pipes
0
2000
4000
6000
8000
10000
12000
2 2 :
2 5
2 3 :
1 2
2 3 :
5 8
0:
44
1:
30
2:
16
3:
03
3:
49
4:
35
5:
21
6:
08
6:
54
7:
40
8:
26
9:
12
9:
59
1 0 :
4 5
1 1 :
3 1
1 2 :
1 7
1 3 :
0 3
1 3 :
5 0
1 4 :
3 6
1 5 :
2 2
1 6 :
0 8
1 6 :
5 4
1 7 :
4 1
1 8 :
2 5
1 9 :
0 1
1 9 :
1 2
1 9 :
1 8
1 9 :
2 5
1 9 :
3 1
1 9 :
3 8
1 9 :
4 5
1 9 :
5 0
1 9 :
5 3
1 9 :
5 6
1 9 :
5 9
2 0 :
0 3
2 0 :
0 8
2 0 :
1 1
2 0 :
1 6
2 0 :
1 9
2 0 :
2 2
2 0 :
2 5
2 0 :
2 9
2 0 :
3 2
2 0 :
3 5
2 0 :
3 8
Time
Total sum of all pipes Feed rate
Mass flow through the pipes
0
500
1000
1500
2000
2500
2
3
:
3
4
0
:
2
9
1
:
2
4
2
:
1
4
3
:
0
2
3
:
4
9
4
:
3
8
5
:
5
5
7
:
1
0
8
:
3
3
9
:
2
2
1
0
:
1
7
1
1
:
0
9
1
1
:
5
7
1
2
:
4
7
1
3
:
4
4
1
4
:
4
5
1
5
:
4
5
1
6
:
4
2
1
7
:
4
1
1
8
:
3
0
1
9
:
1
9
2
0
:
2
2
2
1
:
3
1
Pipe 1 Pipe 2 Pipe 3 Pipe 4 Pipe 5 Pipe 6 Pipe 7 Pipe 8
Figure VII
Feed Rate / Sum of All Pipes
0
2000
4000
6000
8000
10000
12000
2
3
:
3
4
0
:
3
1
1
:
2
8
2
:
1
9
3
:
0
9
3
:
5
7
4
:
4
8
6
:
0
9
7
:
3
4
8
:
4
9
9
:
4
1
1
0
:
3
7
1
1
:
2
8
1
2
:
1
9
1
3
:
1
1
1
4
:
1
1
1
5
:
2
2
1
6
:
1
9
1
7
:
1
9
1
8
:
1
4
1
9
:
0
5
1
9
:
5
8
2
1
:
1
2
Feed rate Sum of all pipes