An Update on Consistency Transmitter

Measurement Technology & Process

Considerations
Mike Hendricks
Applications Manager
BTG Americas Inc.
Agenda

Consistency Overview

Rules for Good Control

Dilution

The Control Loop

Consistency Transmitters

Calibration
Consistency Overview
 Definition of pulp consistency

The weight of suspended solids per litre of liquid

Units used:
1 % Cs = 10 g/l
1 g/l = 1000 mg/l
 Suspended Solids

Consist of
Fiber, Fines, Fillers

Fiber comes from different

processes:
Chemical
Sulphate/Sulphite
Mechanical
Ground wood
CTMP/TMP
Recycled
Why measure pulp consistency?

Most critical papermaking parameter

Controls basis weight of paper

Optimum performance of process

equipment

Mass flow calculations

Maintain ratio between fiber types

Dosing of additives
 Or put another way

Improve the bottom line

Reduce fiber usage
Optimize fiber blending ratio
Reduce energy usage
Reduce chemical usage
Improved process stability

There are no Silver Bullets in process control

It all starts with consistency and flow control
Everything else builds off of and further improves control
Advanced Process Control

Chemicals are added as percent

applied (flow of chemical x
concentration) / (flow of stock x
consistency)
If you start with bad production rate
(flow of stock x consistency)
information there is no amount of
APC that will make up for it
In bleach plants (and other areas) it
is human nature to over apply
chemicals due to the importance of
maintaining pulp quality
requirements
 Fiber Blending
Existing CT StDev New CT StdDev
0.15 0.10 Directions: Fill in yellow cells

Mill & Location

Application
Setpoint (%Cs) 3.5
Production rate (tpd) 100
Fiber cost a ($) $800.00
Fiber cost b ($) $500.00
Ratio a 25%
Ratio b 75%
Run days per year 350

Existing CT Existing CT
Fiber A Fiber B
Consistency Flow TPD Fiber Cost Per Day Consistency Flow TPD Fiber Cost Per Day
3.200 119 22.86 $18,285.71 3.200 358 68.57 $34,285.71
3.500 119 25.00 $20,000.00 3.500 358 75.00 $37,500.00
3.800 119 27.14 $21,714.29 3.800 358 81.43 $40,714.29
2.143 $1,714.29 6.429 $3,214.29

New CT New CT
Consistency Flow TPD Fiber Cost Per Day Consistency Flow TPD Fiber Cost Per Day
3.300 119 23.57 $18,857.14 3.300 358 70.71 $35,357.14
3.500 119 25.00 $20,000.00 3.500 358 75.00 $37,500.00
3.700 119 26.43 $21,142.86 3.700 358 79.29 $39,642.86
1.429 $1,142.86 4.286 $2,142.86

StDev in existing CT over uses fiber A by (tons per day): 0.714 StDev in existing CT over uses fiber A by (tons per day): 2.143
Excess fiber cost per day: $571.43 Excess fiber cost per day: $1,071.43
Excess fiber cost per year: $200,000.00 Excess fiber cost per year: $375,000.00

Total excess fiber cost $575,000.00

Benefits of consistency control

Adjust to optimum consistency:

Process plant efficiency ($ & Tonnage)
Final product specifications
Basis weight
Formation
Physical strengths

Remove process variations:

Optimum equipment performance
Runnability
MD & CD variations
 Process equipment examples

Stock prep refining 3-5%

Screening & cleaning <1%
ClO2 Bleaching 10-16%
Flotation cells <1%
Machine chest 3-4%
RF Pulping 3-6%
Headbox <1%
Whitewater <0.4%
HD storage 10-16%
Pulp Consistency Ranges
Illustration of Consistency
Rules for Good Consistency
Control
Rules for Good Consistency Control

Dilution steps should be small:

The consistency should be reduced by no
more than 20 - 25% in one step

Ensure good mixing:

Correct mixing of the suspended solids and
the water is essential
Design the system around the pump to take
advantage of the pumps mixing ability
Rules for Good Consistency Control

Ensure good dilution:

The time required for the dilution to reach the pump
should be short

The dilution water flow must not interfere with the

free flow of stock to the pump

Dilution header pressure must be constant

Use a dilution control valve with a high resolution and

suitable range
Rules for Good Consistency Control

Minimize Dead Time (transmitter type and

position):
2-3 sec Short
6 sec Normal
15 sec Too long

Ensure good chest agitation:

The pump will mix stock and water when they arrive at
the impeller together. The pump cant smooth out
stock variations which are strung out along the suction
line, which is why chest consistency variations travel
Rules for Good Consistency Control
Locate consistency transmitter correctly:
Follow the manufactures instructions for the piping
run for the transmitter
Distances, velocities, pipe angles

Install consistency transmitter correctly:

Check if turbulent or plug flow is essential for the
transmitter function
Mount the transmitter according to the
manufactures recommendations
e.g., Mounting is critical for blade transmitters which must
be parallel to the stock in both planes
 Single Dilution

Cs = 4%
Max 20-25 %
dilution in one
stage

Cs = 5%
Multiple Dilution Steps
 Double Dilution

Coarse Dilution
(75-80%)
(8-12% 5-6%)

Fine Dilution
(20-25%)
(5-6% 4%)
The Control Loop
 Glossary

Cs consistency
CT consistency transmitter
FT flow transmitter
PV process variable
SP setpoint
DCS distributed control system
CE control element
Repeatability how well multiple measurements of the
same thing are repeated
Accuracy the sum of CT and lab repeatability
How consistency control works

Consistency of the stock in the

Set point
SP
pipe is measured (process
variable or PV)

PV is compared to a set point

(SP) in the DCS (control system)

The output from the controller

affects an actuator on a dilution
water valve (control element or
CE)

The amount of dilution water is

varied to alter the consistency
to the set point
 Consistency Control Loop

1. Stock chest with agitator

2. Stock pump
3. Dilution water supply
4. Dilution control valve
5. Dilution nozzle
6. Consistency transmitter
7. Controller/DCS
8. Sample valve
9. Lab work

Everything is important

The weakest link will decide the result!

 Control Parameters

PID Constants
Proportional
Integer
Derivative
Dead Time
Filter
Sample Time
Update Rate
Dilution Valve "PID en updated feedback" by TravTigerEE - Own work. Licensed
under CC BY-SA 3.0 via Commons -
https://commons.wikimedia.org/wiki/File:PID_en_updated_fee
dback.svg#/media/File:PID_en_updated_feedback.svg
 Constants

Proportional current error

Integer accumulation of past error

Derivative predicted error normally not used

in Cs control
 Dead Time

Should be minimized as much as possible

Make A B as short as possible

Variations with frequency < to the dead

time are not possible to control at all
B
Variations with frequency < 12 times the
dead time are hard to control effectively
A
 Dead Time

Dilution valve closing

Td 4
Deadtime ~ 20 sec Open loop Time Constant () ~ 30 sec
63%
Output signal blade type
transmitter (not in control)
 Filter

Time constant is 63% of full change

Filter will cause a slower response and increase dead time

High frequency noise disturbs good control

Rule of thumb is to have a time constant on the filter less than

0.5 times process time constant

Process time constant = 30 sec

Filter time < 15 sec
 Update Rate
Todays transmitters are digital

They update the output signal after each

measurement

The update rate has to be at least the same or faster

than the controllers sample time

5-10 times faster than the time constant of the

controlled process

Process time constant = 30 sec

Update rate every < 3 sec
Dilution
 Dilution
Dilution water
Pressure controlled
Adequate pressure (+0.5bar)
Isolated
Control valves
Properly sized
Maintained properly
Stiction
Actuators
Located close to line
 Dilution point

V1 = 3-5 m/s
V2 = 0.5-5 m/s
G = Min 15-50 mm
Delta P = min. 0.5 bar
L = 3 x D / min 1 m
 Dilution nozzle
The unfortunate fact of the matter is that we are
stuck with what we got
There is a right way
Consistency Transmitters
There is no consistency transmitter
All transmitters are
inferential in that they
measure something other
then consistency and infer
consistency based on
calibration (correlation) with
laboratory analyze of a
representative sample or
samples

The laboratory is the only

real measurement, and
while there is a TAPPI
standard, not everyone follows
it, lab repeatability
 Repeatability vs Accuracy

Repeatability how well something measures the

same thing over and over, i.e. lab and transmitter
results are repeatable

Accuracy what you get when you put lab and

transmitter repeatability together
Comparing technologies
 Velocity and consistency profiles

What does the

process provide?

What does the

transmitter need
 Shear force

Shear force

Most widely used

Measures only the fiber part of the suspension

Needs fiber network (+1.5% Cs)

Insensitive to color and chemicals

Static blade
Low cost

Inexpensive installation

Flow dependent

Require plug flow - dead time issues

Lowest repeatability

1.5-16% cons
 Static Blade Flow Changes
Long Fiber from Fractionator
8000 gpm
3.0% Consist

Increased Flow increases consistency reading

Decreased Flow lowers consistency reading
1.5% Consist
Adds variability into refiners
4000 gpm

Green: Stock Flow (gpm)

White: % Consistency
43
Measuring principle

A A>B
 Moving blade

Low cost
Inexpensive installation
Less flow dependent
Require plug flow - dead time
Lowest repeatability
1.5-8% cons
Rotating shear force
Virtually flow insensitive

High repeatability 0.003%

Large range of applications

More expensive

Expensive installation
Measurement principle
 Rotating shear force
1.5-16% consistency

Used at high accuracy demand/difficult applications

Can be used most anywhere in the process

many variants / proven safe choice

Little effect from flow, fiber composition and type, freeness

etc.

Not affected by air, pressure, conductivity, color, etc.

Weak points: More moving parts, some influence by fiber

type, does not measure total consistency, installed cost
Transmitted optical
True total consistency

Low consistency, 0.01 3%

Measures fine particle and fiber separately

Large range of low consistency applications

Limited consistency range up to ~3.0%

 Transmitted optical
DC signal The Peak

Both small and large No fiber passes the light

The signals particles pass the light gap gap
combined provide the A fluctuating signal Transmits a large amount
basis for evaluating (DC signal) of light
the consistency of the Information on both small The Peak
pulp suspension and large particles in the Information on the filler
stock content of the stock
 Transmitted Optical
0.01-3/3.5% Consistency

Virtually insensitive to fiber type and composition,

fillers, flow, color, etc.

Simple design - no moving parts - low maintenance

Low installation cost

Suitable for low-consistency applications: screens,

cyclones, de-inking cell, effluents, etc

Weak points: High consistency, +25% filler content,

minimum of 1.3m/s flow velocity
 Reflected optical

True total consistency

Wide range, 1.0 16%
Measures total consistency
Large number of application possibilities
Installed in turbulent flow minimum process
deadtime
Inexpensive CT
Inexpensive installation
No moving parts
Conductivity & pressure non-issues
Small measuring volume, demands good mix of
the suspension (turbidity)
TiO2 & changes vivid colors
 Reflected optical
DC signal The Peak DC and Peak are used
for the total
consistency.
Min. gives information
on fines and fillers

Spread

Min. and Peak

The signals
combined provide the
basis for evaluating
the consistency of the
pulp suspension
Both small and large
particles pass the light
beam
A fluctuating signal
(DC signal)
Information on both small
and large particles in the
stock
One single fiber passes
the light beam
Reflects a large amount
of light
The Peak
Information on the fiber
content of the stock
Microwave Transmitter

+ Accurate
Close to total consistency
Measures both fiber, fines and
filler
Not sensitive to fiber shape and
size

Air sensitive
Conductivity, temperature and
vibration sensitive
Limited range of application
 Microwave Transmitter
0-8% consistency

Good repeatability

Application specific typically blend or machine chest

Virtually unaffected by flow, fiber composition and type,

freeness, etc. With compensation can measure total
consistency

No moving parts - low maintenance

Weak points: High sensitivity for entrained air/pressure,

temperature and conductivity variation. Pulp must be
screened
 Selecting the right transmitter

Consistency range Line pressure

Accuracy needs Conductivity
Chemical application Chemicals used
Basis weight Metallurgy
Fiber blending pH
Specific energy Flow/Velocity
Process area Available calming
Pulp type distance/process deadtime
Recycle
Virgin
Synthetic
Mixed
 Measurement versus application

Application Accuracy Needed Impact of 0.5% inaccuracy

Broke Consistency Low Can be high for blending control
Refining consistency High Energy overuse
Blending control Medium Furnish cost differential
Machine Chest High Poor Basis weight control
Type Accuracy (Std. Dev.) Flow sensitivity Entrained air sensitivity
Static blade Low Poor Medium
Moving Blade Medium Medium Good
Microwave Medium-High Good Poor
Active blade Medium-High Medium-Good Good
Optical High Good-Medium Good
Rotary, Mechanical High Good Good
 Best Practice
Appropriate
Sensor Correct
Sample Good Stock
Valve mixing
Correct Correct
Calibration Dilution
Valve

Sensor in
right place

Control Max 20%

loop dilution in 1
Tuning step
 Installation guidelines

Rotating 3x & 1.5x (<8%) or 1.5x & 1.5x (>8%)

Blades 10x & 5x
Optical immediately after pump
Calibration

Or how to ruin a
perfectly good
day
 Calibration

Guess what they arent all the same!

Shear force is non-linear
Optical is linear
Microwave is linear
Filler & Fines effect
Transmitter repeatability
Also.
A good calibration requires samples from all grades
A good calibration requires samples at different consistencies
Dont short change it!!
 Different calibrations - why
Different calibrations
Single point
Multipoint
Recipe curves

How to select
Application
Quality of lab
Process
 Calibration ideas
The process determines the calibration
You cant force the calibration
Good single point calibration will work
Good multipoint calibration will work

Questionable lab
Single point calibration
Recipe calibration
 Calibration thoughts..
Application
Machine chest
Pulper
Dump chest
Feed to refiner
Grade structure
Fiber mix
Number of samples
How much can operations move the dilution control
valve
Single point calibration
Recipe calibration
 Single Point Calibration Data

Single Point
Lab Cs Raw Signal

Single Point Calibration

80
4 55 70

60
Raw Signal

50

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Lab Cs
 2 Point Calibration

2 point
Lab Cs Raw Signal

3 40
2 Point Calibration

4.5 65 80

70

60
Raw Signal

50

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Lab Cs
 Multipoint Calibration Data

Multipoint
Lab Cs Raw Signal
0 24
3 40
Multipoint Calibration
3.5 45
4 55 80

4.5 65 70

60
Raw Signal

50

40

30

20

10

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Lab Cs
 Combined Calibration Data

Single Point 2 point Multipoint

Lab Cs Raw Signal Lab Cs Raw Signal Lab Cs Raw Signal
0 24
3 40 3 40
3.5 45
4 55 4 55
4.5 65 4.5 65
 Calibration Comparison

Combined

80
70
60
Raw Signal

50
40

30
20
10
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Lab Cs

Multipoint Single Point 2 Point

Accuracy of Sample

Sampling method

Sampling technique

Lab equipment

Lab repeatability

Transmitter repeatability
 Lab

Procedures
TAPPI
A repeatable procedure
Does everyone uses same procedure

Equipment
What is and isn't there
Condition of equipment
Scale significant digits, 2? 4?

Personnel
Trained
Does anyone care
 Ball valves

Sample is taken at pipe

Dewaters pulp unless fully opened

Everyone opens it differently

Gives large volume and splashes at high

line pressure

Low price

Exception?
 Pulp Sampling

Important to have an accurate and repeatable sample

Well-designed valve:
Takes the sample well inside the pipe wall
Always fully open or closed
Not dependent on operator
Safe

Selecting a sample valve

Consistency
Pressure
Error on side of safety & reliability Water film
 Calibration
See. It isnt that bad
 Success
Management
Drives the level of importance
Provides the capital
Understand and support reliability program

Operations
Understand the benefits of running in control
If the control loop isnt working they investigate
Ownership of their process area
 Success
Maintenance
Reliability program
calibration verification
Transmitter inspection
Control loop

Labs
Good sample valves
Standard procedures that everyone follows,
Good lab equipment
SPC or control chart documentation
Summary

Consistency Overview

Rules for Good Control

The Control Loop

Dilution Point

Consistency Transmitters

Calibration
 Thank You

Questions??

Mike Hendricks
Application Manager
BTG Americas Inc.
715-550-8158
mike.hendricks@btg.com