Mechanism of Refining - TMP Low Consistency

Download as pdf or txt
Download as pdf or txt
You are on page 1of 71

Mechanisms in the refining

zone for development of


physical properties of TMP
fibres in a low-consistency
refiner
Gap clearance and vibration measurements

Oddbjørn Eriksen and Bertil Roos

2005

According to Innventia Confidentiality Policy this report is public since 2011-02-04


Mechanisms in the refining
zone for development of
physical properties of TMP
fibres in a low-consistency
refiner
Gap clearance and vibration measurements

Oddbjørn Eriksen and Bertil Roos

Report no. 115, 2005

Cluster:
Mechanical pulp research cluster

Distribution restricted to:


Eka Chemicals, Holmen Paper, Norske Skog, Stora Enso, Södra Cell

a report from STFI-Packforsk


According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

2 (41)

Acknowledgements
This study has been performed as a part of the project LC refining mechanisms within the
cluster research program of Mechanical pulp 2003 – 2005. The industrial members of this
cluster through the companies Eka Chemicals, Holmen Paper, Norske Skog, Stora Enso
and Södra Cell are gratefully acknowledged.
A special recognition is given to the companies GL&V and J&L Fiber Services for
providing the refiner plates used in the trials at the EuroFEX pilot plant facilities. Dametric
AB is also acknowledged for providing the gap clearance sensors. A great thank to the staff
at the EuroFEX pilot plant. Börje Svensson as the excellent refiner operator is gratefully
acknowledged for his patience with the researcher’s endless tripping around in the pilot
mill.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

Table of contents
Page
1 Summary..................................................................................................... 4
2 Introduction ................................................................................................ 6
3 Background ................................................................................................ 7
3.1 Gap clearance measurements ..................................................................... 7
3.2 Vibration measurements .............................................................................. 8

4 Material and methods .............................................................................. 10


4.1 Refiner........................................................................................................ 10
4.2 Running conditions..................................................................................... 11
4.2.1 Trial 1 ...................................................................................................................... 12
4.2.2 Trial 2 ...................................................................................................................... 13
4.2.3 Trial 3 ...................................................................................................................... 15
4.3 Process measurements ............................................................................. 16
4.3.1 Vibrations ................................................................................................................ 16
4.3.2 Gap clearance......................................................................................................... 17
4.4 Data analysis methods............................................................................... 22

5 Results and interpretations..................................................................... 23


5.1 Trial 1 ......................................................................................................... 23
5.1.1 Gap measurements ................................................................................................ 23
5.1.2 Vibration measurements......................................................................................... 27
5.2 Trial 2 ......................................................................................................... 30
5.2.1 Gap measurements – DCS data............................................................................. 30
5.2.2 Gap and vibration measurements........................................................................... 31
5.2.3 Vibration frequencies .............................................................................................. 33
5.3 Trial 3 ......................................................................................................... 34
5.3.1 Gap measurements ................................................................................................ 35
5.3.2 Gap and vibration measurements........................................................................... 35

6 Summary................................................................................................... 37
7 References................................................................................................ 39
Appendix A:
STFI-Packforsk EuroFEX Laboratory Report. Trial 1 – Week 34, 2004 .............. 8 p.

Appendix B:
STFI-Packforsk EuroFEX Laboratory Report. Trial 2 – Week 41, 2004 .............. 8 p.

Appendix C:
STFI-Packforsk EuroFEX Laboratory Report. Trial 3 – Week 22, 2005 .............. 8 p.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

4 (41)

1 Summary
This technical report addresses the investigations of high-frequency vibration
measurements of the refiner housing and measurements of the plate clearances in the
refining zone of a low-consistency disc refiner during refining of mechanical pulps. The
report shows results from three experimental trials in the Beloit DD 4000 refiner at the
STFI-Packforsk EuroFEX laboratory in Stockholm. Five different plate sets have been
tested at various process conditions. The measurements were accomplished with
prototypes adjustable gap clearance sensors (AGS) provided by Dametric AB as well as an
external Brüel & Kjær accelerometer attached to the refiner housing. Additional process
variables were accessible from the distributed control system (DCS) of the pulping line.
The present study has revealed some features of the Beloit DD refiner as well as the
different refiner plates investigated. Among the most important it is shown that the twin-
concept with two refining zones and a floating rotor disc gave not always similar refining
conditions in the two refining zones. It is reasonable to questioning the equality of the
running conditions in this refiner due to an observed asymmetrical position of the rotor
disc. It is strongly indicated that the movement of the rotor disc was uneven in some runs.
Thus, it is reasonable to question the homogeneity of the double disc concept and evaluate
whether an additional dimension regarding the uncertainty is introduces compared to single
disc operation.
It can be concluded that a considerable advantage of the AGS-sensor compared to external
disc displacement sensors is that the zero-point setting can be checked and adjusted when
the refiner is running. The uncertainty in zero-point setting is much less for the AGS
system compared to what can be achieved with the original gap displacement sensor. Both
measurements seemed to give an accurate measure of the shift in gap clearance during
operation.

Vibration measurements are believed to show a fingerprint of the refining behaviour in the
refining zone. Frequency analysis of the vibration signals was used to identify the
movement of the rotor bars at different radial positions of the refiner plates. The
relationship between the frequency peaks and the plate pattern was unambiguous. Mainly,
the frequencies associated to the movement of the bars located close to the periphery of
the plates dominated the spectra. Sideband frequencies were also regularly observed. The
four radial directed open spaces from the inlet to the periphery of each of the refiner plates
also called flow channels influenced the flow of pulp through the refining zone. Their
periodical influence appeared commonly in the frequency spectra as sideband frequencies.
Increased speed of the refiner gave significantly higher vibrations. The plates with the
coarse bars incorporated dams gave the highest vibration level probably due to the dams
ability to obstructed the pulp flow through the grooves and force the pulp into the plate
gap.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

A relationship between the sum of gap clearance measures and the overall vibration level is
proposed. The vibration level reached an optimum at a gap clearance of approximately
200 µm. It is conceivable that the gap clearance should reach an adequate level regardless
of the flow rate. However, it is still hypothetical if this level can be associated with the
observed optimum of the gap and vibration relationship.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

6 (41)

2 Introduction
This technical report shows results from the three experimental trials in the Beloit double
disc refiner at the STFI-Packforsk EuroFEX laboratory in Stockholm. Five different plate
sets have been tested at low-consistency conditions. The measurements are collected
during different operational conditions. This report addresses the investigations of high-
frequency vibration measurements of the refiner housing and measurements of the plate
clearances in the refining zone. The latter is associated to the use of modified TDC sensors
(True Disc Clearance) also called AGS - adjustable gap sensor. Mainly, the experiments
undertaken by refining of softwood (Norway spruce) mechanical pulp are treated in this
report.

The work has been a part of the project LC refining mechanisms of the 2003-2005
mechanical pulp research cluster program. Additional results from this study are covered in
other sub-reports by Hammar (2005) and Eriksen (2005). The latter report covers an
analysis of the measures of rapid pressure pulses in the refining zone. That work as well as
the plate clearance measurements reported in the present study has been conducted as a
co-operation between two different projects in two different research clusters at STFI-
Packforsk. The co-operation has been accomplished through the projects LC refining
mechanisms Refining fundamentals and Refining fundamentals in the Mechanical pulp
research and Advanced fibre management (AFM) cluster programs, respectively.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

3 Background
A summary of the prevailing knowledge of the refining of mechanical pulps in low-
consistency refiners is reported by Hammar (2004). Furthermore, a literature survey of
measurements in the refining zone of low-consistency refiners is written as a part of the
present project (Eriksen and Hammar 2005).

Low-consistency refining is primarily applied as post or rejects refining in the mechanical


pulping process. The work at STFI-Packforsk has put more attention on the LC refining as
an earlier stage in the mechanical pulping line. This is done in order to reduce the specific
energy consumption without loosing the highly desired pulp quality features of high-
consistency pulps. However, the operation of the LC refiner is more demanding due to a
narrower operating window with a plate gap down to a few fibre widths. The plate
clearance in a high-consistency refiner is several times larger. The need to control the
process operation is essential in order to create a pulp of desired quality. Thus, it is
necessary to collect more information from the process. The topical information of interest
is associated to the flow phenomena and the fibre coverage of the bars in the refining zone
as well as the stresses applied to the fibres in contact with the bars. Such information can
be obtained using sensors and measurement technology. The most relevant variables to be
analyzed are probably force, pressure, temperature, residence time, vibrations and the
quantity of pulp in the refining zone. In addition, visualization of the pulp flow behaviour
using photographical methods will support the interpretation and increase the overall
comprehension. In general, there is limited information about the flow performance of the
pulp in the refining zone, and thus, the knowledge of the fundamental refining mechanisms
are lacking.

3.1 Gap clearance measurements


Nordman and co-workers (1981) were pioneers when they conducted plate clearance
measurements in LC refiners using an eddy-current sensor. The size of the sensor was of
such dimension that it was necessary to modify a small part of the rotor disc in opposite
position of the sensor. This was done in order to ensure a sufficient accuracy of the gap
measures. A smooth surface larger than one single bar width was required. The initial
accuracy of the measurements was stated to be 1 µm, however, the reliability of the
absolute values was observed to be 20 times larger. Their experiments were conducted such
that the span of the bar clearance measures varied between 10 and 400 µm. Nordman and
co-workers found that it was challenging to establish a real zero-point. They measured the
zero-point as a post-operation after each refining trial by forcing the rotor disc towards the
stator until metallic contact was obtained.

A main conclusion of the work by Nordman et al. (1981) was that the plate gap clearance
was especially sensitive to changes in the rotational speed of the refiner. Increasing the
angular velocity at a constant beating load caused the bar clearance to increase, and as
STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

8 (41)

expected, increasing the beating load resulted in decreasing plate clearance. However,
Nordman and his co-workers could not establish a relationship between the bar clearance
and the pulp quality. They stated nevertheless that the gap mainly was too high to assume
that single fibres could be trapped between the bars on the opposite discs. Thus, they
concluded that if fibres receive some kind of compressive mechanical treatment in refining
then the fibres appear in bundles or flocs of fibres. On the other hand, they found that the
fibre dimension (hardwood vs. softwood) and the physical conditions of the fibres
determined the gap clearance.

Disc clearance sensors are more or less established technology in high-consistency refining.
Today, relative gap changes can be measured. However, still there are a lot of uncertainties
associated to gradually wear and offset values. The foremost studies connected to gap
measurements in mill-scale refiners were done 25 years ago. Stationwala et al. (1979)
reported from a mill-study where the plate gap clearance was measured by eddy-current,
non-contact displacement type proximity transducers. The reproducibility of the measured
values during ordinary operating conditions was reported to be in the order of ± 40 µm.
Thereafter, Dahlqvist and Ferrari (1981) and Jackson et al. (1986) introduced mill-scale
studies using Sunds Defibrator's true disc clearance (TDC) sensor for supervision of the
refiner plate gap and control of the motor load. As a counterpart to the present study,
Mohlin (2006a, b) reported results from pilot-scale studies with LC refiners where the
adjustable gap sensors (AGS) were tested during refining of chemical pulps. The results
seem to be very promising. Mohlin stated that the plate clearance is a favourable variable to
compare refining results compared to the theoretical term refining intensity.

3.2 Vibration measurements


Only a few studies are previously reported about vibration measurements of the refiner
housing. Pettersen and Gunstrøm (1980) and later Pettersen (1986) reported from a study
on both conical as well as disc refiners. They used phonometric measurements and
accelerometers on the refiner housing in order to correlate the noise and vibration
measurements to the motor load of the refiner. Among the results, it was shown that the
pressure pulses generated from bar crossings created vibrations both in the stator and the
rotor discs, which in turn generated acoustic oscillations. The analyses revealed that the
corresponding peak frequencies were shifted towards lower frequencies when the motor
load increased. However, the shift was only a few Hertz. This phenomenon was assumed
to be due to lag in the motor related to its synchronized rotational speed.
Additional results from this study shown that new plates generated more noise and
vibrations than worn plates, and that higher axial thrust gave increased vibration
amplitudes. In the latter case the peak frequencies associated to the inlet region were more
dominating compared to the higher frequencies, associated to the outer part of the refining
zone, which were predominant at lower axial thrust. Other observations were that vibration
analysis related to the radial position showed a parabolic curve indicating a parabolic
According to Innventia Confidentiality Policy this report is public since 2011-02-04

distribution of the applied energy. Wave propagation in the surface structure of the refiner
housing were also detected when the housing was exerted in vibrations. It was claimed that
that the clearance between the stator and rotor housing of a conical refiner could vary up
to 20 µm during operation.

Strand and Hartler (1985) and Strand and Mokvist (1987) studied the vibrations of a high-
consistency refiner. They claimed that the origin of vibrations is directly related to the
intensity of the pressure pulses resulting from the passing of rotor and stator bars. Thus,
Strand and Mokvist (1987) assumed that the intensity of the vibrations, which in turn
correspond to the refiner bar frequencies, could be used to characterize the dynamic forces
applied to the fibres. They showed two frequency spectra that strongly indicated that there
were connections between the vibrations recorded by accelerometers attached to the
refiner housing and the refiner plate pattern (bar intersecting area). The frequencies
associated to the accelerometer signal could be divided into four definite regions related to
the breaker bar, coarse bar and fine bar sections in addition to vibrational overtones as are
equivalent termed harmonics (Strand and Hartler 1985).
Eriksen and co-workers (2003, 2006) enhanced the impression of the proposed
interrelation between the pressure pulses created by the overlapping bars and the vibrations
of the housing of high-consistency refiners. The frequency analysis of the pressure
measures in the refining zone gave almost a mirror image of the vibration spectra appearing
from readings from an accelerometer attached on the refiner housing. Thus, it could be
stated that the pressure pulses appearing from the squeezing of pulp between the stator
and rotor bars propagated as vibrations in the mechanical structure of the refiner. Eriksen
and co-workers made also a hypothesis that assumes that the pulp flow propagates radially
outwards in waves caused by low-frequency disturbances as feeding variations and
interruptions by the ribbon feeder or coarse bars.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

10 (41)

4 Material and methods

4.1 Refiner
The refiner used in the present trials was a 600 mm (24 inches) Beloit Jones Double-D
refiner in the series 4000 manufactured by Beloit-Walmsley Ltd. The refiner motor is
frequency controlled in order to allow variable speed. The refiner is driven by a 322 kW
DC motor denoted DMA 315 S 35V manufactured by ABB Inc.

Figure 3.1. The sketch shows the Beloit Jones DD 4000 refiner (Source: Booklet from
Beloit Corp., MA, USA).

The refiner has two flat stationary discs and a double-sided rotor disc in between. The
rotor disc is movable on a spline shaft. The stator discs on the tail-end of the refiner is
movable by a frequency controlled motor, which sets a mechanical load in the desired disc
position. The rotor is assumed to be equally balanced and positioned in the middle between
the stator discs by the input flow of pulp suspension. However, this is not generally
observed. It is assumed that the friction between the spline on the rotor disc and the spline
shaft prevent the equalizing of the position of the rotor disc.

The refining system has one inlet tube, shown in Figure 3.1, and one outlet tube. In the
present and previous trials the refiner has been operated as a single pass refiner. A rebuild
According to Innventia Confidentiality Policy this report is public since 2011-02-04

of the system which was operative from November 2005, allows up to 50 per cent
recirculation flow into the refiner. Moreover, the refiner system is equipped with flow and
pressure controls. In addition, temperatures of the pulp suspension in and out of the
refiner are measured and recorded together with other process variables in a distributed
control system (DCS) provided by ABB.

Figure 3.2. The photography shows the Beloit DD refiner seen from the fixed stator side.
(Source: Ola Johansson, J&L Fiber Services Inc.).

4.2 Running conditions


The three different trials were conducted in the same refiner at conditions that did not
differ too much. However, the main difference between the trials was the use of different
refiner plates. Two of five plate sets were provided by GL&V, while the three others were
delivered by J&L Fiber Services. The refining series were accomplished with first or second
stage high-consistency pulps. The mechanical pulps were collected in big bags at high-
consistency conditions at the actual mills. Water was heated to 95 ºC prior to the dilution
of the high-consistency pulp. The heating of water was accomplished separately in order to
reduce negative influence of the fibres during a long heating process. The consistency of
the diluted pulp was initially set to 3.5 %. The stock was stirred in 20 minutes to
disintegrate the fibres prior to the refining. The refining was conducted as a one-pass run
through pumping of the pulp from one storage chest to another. Samples of the pulp were
taken out from a sample-valve during the refining. Figure 3.3 shows the LC refiner loop of
the EuroFEX pilot mill.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

12 (41)

Sample
A B valve Heat
exchanger

Pump Refiner

Figure 3.3. LC refining loop of the EuroFEX pilot mill (Hammar 2005).

4.2.1 Trial 1
Two different mechanical pulps were LC refined at approximately 3.5 % consistency. Both
pulps were TMP from a Swedish mill. One of the pulps was a first stage high freeness pulp
(450 ml) whereas the other was a second stage pulp of lower freeness (150 ml). The
refining intensity in the LC refiner was initially determined by a selected range of specific
edge loads (SEL) as shown in Table 3.1. The corresponding specific energy consumptions
were controlled by the flow rate through the refiner. Additional information about the
running conditions is enclosed in Appendix A.

Table 3.1. Conditions in the LC refining of trial 1.


Consistency [%] ∼3.5
Temperature [°C] ∼90
Rotational speed [rpm] 700
Edge load Bs [Ws/m] 0.3, 0.5, 0.7 and 0.9
Edge length Ls [km/rev] 8.4 and 15.7

The plate designs from GL&V are shown in Figure 3.4. The plates are termed medium and
fine bars pattern. The terms characterise the large difference in edge lengths (Ls). The
average angle of the bars was similar for both sets.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

Medium Fine

Bars: 3.2 mm Bars: 2.4 mm


Grooves 3.2 mm Grooves 2.4 mm
Depth: 7.1 mm Depth: 6.4 mm
Average angle: 23.8° Average angle: 23.8º
Ls: 8.4 km/rev Ls: 15.7 km/rev

Figure 3.4. GL&V medium and fine type refiner plates.

4.2.2 Trial 2
High freeness pulp (450 ml) from a first stage high-consistency refiner was collected from
the same mill as delivered the pulp to trial 1. The trial was performed at different specific
edge loads at constant flow rates and rotational speed. Table 3.2 shows the frame of the
running conditions. Additional information about the running conditions is enclosed in
Appendix B.

Table 3.2. Conditions in the LC refining of trial 2.


Consistency [%] ∼3.5
Temperature [°C] ∼90
Rotational speed [rpm] 700 and 900
Flow rate [l/min] 300, 400, 600, 800 and 900
Edge load Bs [Ws/m] 0.3, 0.5, 0.7 and 0.9
Edge length Ls [km/rev] 6.1 and 18.6

The plate designs used in the second trial are shown in Figures 3.5 and 3.6. The first plates
were based on a high number of fine bars. Thus, the edge length was large of some 18.6
km/rev. The second plate set consisted of coarse bars. The corresponding edge length was
only 6.1 km/rev. However, it is notable that the plates had full surface dams.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

14 (41)

Bar width Groove width Depth Angle Ls


Remark
[mm] [mm] [mm] [º] [km/rev]
2.2 2.9 6.7 10.0 18.562 No dams

Figure 3.5. The design of the fine bars refiner plates from J&L Fiber Services.

Bar width Groove width Depth Angle Ls


Remark
[mm] [mm] [mm] [º] [km/rev]
3.6 5.2 7.9 10.0 6.126 Full surface dams

Figure 3.6. The design of the coarse bars refiner plates from J&L Fiber Services.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

4.2.3 Trial 3
A high freeness RTS-pulp was taken out from a first stage high-consistency refiner in a mill
in Sweden. The freeness of this pulp was approximately 335 ml. Also a high-consistency
high-freeness pulp similar to the pulps used in the previous trials was treated in the LC
refiner in this trial. The first stage pulps were refined at low-consistency in one passage in
the pilot refiner. The LC refining was carried out at different edge loads and specific energy
inputs at constant flow rates. Table 3.3 shows the running conditions. The lowest flow rate
was omitted when the RTS pulp was run. Additional information about the running
conditions is enclosed in Appendix C.

Table 3.3. Conditions in the LC refining of trial 3.


Pulp consistency [ %] ∼3.5
Temperature [°C] ∼90
Speed [rpm] 700
Flow rate [l/min] (150), 200, 300, 400 and 600
Edge load Bs [Ws/m] 0.5, 0.7 and 0.9
Edge length Ls [km/rev] 8.54

The refiner plates used in trial 3 were also provided by J&L Fiber Services Inc. Figure 3.7
shows the plate design. The plate pattern had some similarities to the medium type
segments as used in the first trial. However, the average angle of the bars of the present
plates was smaller compared to the refiner plates manufactured by GL&V.

Bar width Groove width Depth Angle


Ls Remark
[mm] [mm] [mm] [º]
3.0 4.4 7.3 7.5 8.54 No dams

Figure 3.7. The design of the medium bars refiner plates from J&L Fiber Services.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

16 (41)

4.3 Process measurements


The refining mill at the EuroFEX pilot plant is well fitted with measurement technology
and an updated distributed control system (DCS) from ABB. The prevailing process
measurements are captured by the DCS by a typical frequency of 1 hertz (one sample per
second). Actual process variables, as the flow rate, motor load, gap displacement, and the
temperature and pressure in the inlet and outlet tube, were recorded. Also the gap clearance
measures were transmitted to the DCS. However, a separate data acquisition unit termed
PFI data logger was used in order to collect high-frequency data from selected variables.
These variables were especially the gap clearances and vibration together with the refining
zone pressures. The results associated to the latter variable are reported by Eriksen (2005).

The data acquisition system consisted of a high-speed DAQ-card from National


Instruments Corp. with the following data:
– Type: NI-6115
– Channels: 4
– Resolution: 12 bits
– Internal gain: 0.2-50
– Sampling rate: max. 10 million samples per second (10Msamples/s) on one channel or
1.25 Msamples/s simultaneous on 4 channels
The high-speed DAQ-card was running on a CompactPCI-bus using a standard PC with a
hard disk capacity of approximately 80 Gbyte and an internal memory of 256 Mbyte.
LabVIEW software from National Instruments was used together with the DAQ-system.

4.3.1 Vibrations
The vibration measurements were performed by the use of an accelerometer attached to
the refiner housing. The accelerometer was manufactured Brüel & Kjær. It was used
together with a corresponding charge amplifier and the high-speed data acquisition unit.
The equipment information is as follows:
– Accelerometer: Brüel & Kjær type 4384, serial No. 1390386
– Circular size: 14 mm in diameter, length of 24.5 mm
– Weigth: 11 grams
– Piezoelectrical, natural frequency: 45 kHz (flat transfer function: 0.1 Hz – 10 kHz)
– Gain below 6 dB: 10-35 kHz
– Charge amplifier: Brüel & Kjær type 2635
According to Innventia Confidentiality Policy this report is public since 2011-02-04

During the measurements, the amplifier was located a short distance of some 2-3 metres
from the accelerometer. The data acquisition system was also located close to the refiner.
The coaxial cable between the amplifier and the data acquisition unit was approximately
5-10 metres long. Mainly, a sample rate of 500 000 samples per second (500 kS/s) was used
to record the vibration signal.
Recordings of the axial vibrations of the refiner housing were made when the
accelerometer was attached to the refiner housing on the movable stator disc. The
accelerometer was attached to the refiner using a magnet and an attachment screw. The
position of the accelerometer was close to the gap clearance sensor.

4.3.2 Gap clearance


The plate clearance measurements were acquired by use of two sensors, located in each of
the stator discs. The sensors were provided by Dametric AB. Their Adjustable Gap Sensor
(AGS) is an automatic on-line self calibrating sensor for refiners. The sensor is a modified
True Disc Clearance (TDC) sensor. The measuring principle of the AGS-sensor is based on
magnetic reluctance measurements. It is combined with a special positioning system to
define the zero-point.

Figure 3.8. The adjustable gap sensor (AGS) looks like this. (Source: Dametric AB).

The gap clearance measuring unit consists of two systems that interact and facilitate
automatic calibration sequences during normal operation. With the option to have
periodically initiated calibration sequences, it is possible to compensate for the continuous
wear of the sensor tip (along with its surrounding segment). This would otherwise mean an
increasing measurement error. Figure 3.9 displays the functional arrangement of the AGS
sensor.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

18 (41)

Gap Clearance (disc gap)


to be measured xD

Step motor Position APO-Ref.point: Vibration


plus gear Sensor mechanical stop picup
position”

AGS-position
(APO): ”True Disc
Zero when Clearance”
sensor-tip is in Sensor
plane with fixed
t
Fixed Fillings
(Segments) Moving Fillings
(Segments)

Figure 3.9. The functional arrangement of the AGS-sensor.

The first system is a distance meter measuring the open space between the surface of the
sensor tip and the surface of the opposing rotating disc. This sensor is operating exactly in
the same way as the so-called True Disc Clearance sensor (TDC), the forerunner of the
AGS-system. The open space distance is indirectly measured as the change in magnetic
reluctance. Thus a presumption for the measurement is that the refiner filling surface has
some magnetic properties. The measurement is also designed to give an averaged distance
value over one revolution of the rotor.
The TDC-sensor is located in a cylindrical tube that can be moved and this moving system
constitutes the second part of the AGS-sensor. The second system is a positioning servo
system that performs controlled movements of the bar (with the TDC-sensor) in its
mounting direction perpendicular to the rotating disc. It is referred to as the APO-system
and consists of a step motor, a gear and an inductive position meter with high accuracy and
high resolution. The position servo uses a mechanical stop position as a reference point at
self calibrations and a parameter called Back Home position (at a defined distance from the
stop position) to define the zero point.
A vibration pickup (an accelerometer) built in at the tip of the sensor bar can be considered
as a special part of the measuring system. The sensing of the vibration level allows
detection where the sensor tip touches the rotating segments. A scheme of the measuring
system for the DD-refiner, which has two sensors, is shown in Figure .
According to Innventia Confidentiality Policy this report is public since 2011-02-04

Fig. 1
Control Unit for AGS

Analog Outputs:
(4-20mA)
2 Gaps & 2 Temp.
AGS-F or
AGS-R
(”fixed”) 2 TDC & 2 Pos.

Vibration signal:
Voltage proportional
to acceleration sensor
signal

Movable Fixed
Stator Stator

Measurement of
Stator Position
Rotating disc, self adjusted
position between the stators

Figure 3.10. Scheme showing the installation of gap sensors in the DD-refiner.

Zero-point determination and the calibration sequence


The calibration of the distance sensor (the TDC-system, including the electronic system for
signal processing) is an automatically performed sequence starting with finding the zero
point as the touch point where the tip is moving gradually closer and closer to the rotor
until it gets approximately direct contact. The contact point, here referred to as the touch
point, is the event when the accelerometer sensor (located in the sensor tip) gives a signal
that exceeds a pre-set level. Immediately after a touch the sensor is moved backwards away
from the fillings to a well-defined distance of 1 mm. After capturing the new scale factor
value, the sensor is moved back to its normal zero position. The changes in zero point and
scale factor are displayed on the AGS menu. It should be noted that during (any)
calibration sequence the actual gap must not be changed. This is automatically checked by
the AGS unit with the aid of a connected analogue signal for the rotor/stator position. If a
change greater than about 0.01 mm is detected, the calibration sequence is automatically
interrupted.
The so-called TVD system (Touch point Vibration Detector System) uses the
accelerometer signal from the vibration pickup to determine when the touch point has
occurred. A general scale factor can be applied for the signal and, furthermore, each of
three frequency domains can be assigned with a weighting factor to adapt the signal to the
expected vibration pattern at the touch. In the STFI-Packforsk pilot scale studies, the same
weighting factor, 1, has been used all the time. The disc clearance signal is in the present
design given as an average for each revolution

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

20 (41)

Experiences to date
Although the AGS system is indeed very complex, only the rotor speed has to be defined
as a running parameter before start. There are e.g. several calibration parameters that need
some experience before they can be set. All system parameters are easily available for
changes with suitable warnings or blockings for avoiding harmful operations when the
machine is running.
With the demands on high accuracy for the gap clearance in LC-refining, much time was
spent with the calibration of the signal. Besides defining the zero point, the non-linearity of
the signal had to be taken into account. For our research trials, the sensor was modified so
that the APO-signal was made available and a calibration curve could be produced for each
setting (October 2004).

v47-M-segm-Rgiv v12-M-segm-Rgiv
v12-F-segm-Rgiv v21-22-spec.segm-Rgiv
1,1
1
0,9
0,8
True gap (mm)

0,7
0,6
0,5
0,4
0,3
0,2
0,1
0
-0,1
-0,1 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 1,1
AGS-calculated gap (mm)

Figure 3.11. The curves show the calibration of the true disc clearance for three different
refiner plates in the DD-refiner.

From logged data it is possible to recalculate the gap values taken during the test.
Figure 3.11 shows calibration curves for fillings from two suppliers. The difference
between one of the refiner plates compared with the other two is probably due to different
magnetic properties. The linearization will be included in the sensor in future systems.

Zero-point setting
According to Innventia Confidentiality Policy this report is public since 2011-02-04

The zero-point setting becomes very important considering the small gap clearances used
in LC-refining. This is especially the case in R&D applications where different refiners and
refiner fillings are to be compared. Most refiners are equipped with an external system for
measuring the movement of the rotor/movable stator. To set the zero point, the discs are
moved together with an unloaded refiner running at low speed. This is very inaccurate.
Figure 3.12 illustrates that the zero point for the external measurement is difficult to define
and that comparing results for different fillings is difficult with only an external (and
thereby only indirect) measurement of the gap clearance. A considerable advantage of the
AGS-sensor is that the zero-point setting can be checked and adjusted when the refiner is
running.

2,5
Sum gap1&2 [mm]

2,0

1,5

1,0

0,5 Coarse
Fine
0,0
0,0 0,5 1,0 1,5 2,0 2,5
Moving stator position [mm]

Figure 3.12. A comparison is shown between the gap clearance measures and external
measurements of the position of the moving stator. Results from DD-
refiner using bleached softwood market pulp. The measured gap is
presented as the sum of the two gaps in the DD-refiner.

However, even with the AGS sensor the setting of the zero point for the gap clearance can
be associated with some uncertainty. The level of the vibration signal used to detect the
touch point has to be defined to a reasonable level, since too high a level might in some
situations lead to a troublesome wear of the tip (and a displaced zero point). Also, any
misalignments in the system or wobbling of the rotor will make it difficult for the AGS
sensor to find the true zero gap clearance at the touch. Since such phenomena exist as well
as displacements due to material expansions with increased temperature, it is not possible
to define a touch situation which corresponds to the true zero gap clearance with a running
machine. It has also been shown that other external factors can contribute to make the
touch event inexact. For example, the level of the flow rate and refining consistency has a
noticeable effect on the zero-point setting at the calibration. When gap clearance
STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

22 (41)

measurements are used in R&D, the calibrations should be done at similar conditions.
However, the uncertainty in zero-point setting is much less for the AGS system compared
to what can be achieved with an external position sensor i.e. original gap displacement
sensor.

4.4 Data analysis methods


The recorded data were analysed using applications in LabVIEW and Matlab as well as
spreadsheet in Microsoft Excel. Data were visually inspected and analysed statistically. Also
comparisons between data recorded by the DCS and the high-speed logger were
conducted. Regular statistical averaging and variance methods were used. However, the
vibrations demanded other analysis methods in addition. Thus, frequency analysis was also
important in this study. The normalized amplitude spectrum analysis performed using FFT
(Fast Fourier Transform) was used to investigate the periodicity of the vibration signal. The
normalized term was used to show the relationship between the amplitudes in the
frequency domain with the corresponding amplitudes in the time domain. A unit sinusoid
in the time domain corresponds to unit amplitude in the frequency domain. The FFT-
routine in Matlab was used in the calculations.
The total number of samples included is determined by 2n. The maximum number of
samples (N) used in one FFT-calculation have been 221 (2097152) samples. The frequency
spectrum produced by this N point FFT consists of (N/2+1) samples equally spaced
between zero and one-half of the sampling frequency. Assuming a sample rate of 500 kS/s
i.e. sampling frequency of 500 kHz, this gives a frequency resolution of approximately
0.25 Hz.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

5 Results and interpretations

5.1 Trial 1

5.1.1 Gap measurements


Figure 4.1 shows data from two different measuring units collected at various process
conditions. Data from the gap sensors were collected simultaneously with the data from
the original disc position device in order to determine the accuracy of the plate gap
clearance measures. The disc position device measures the distance between the stator discs
subtracted the width of the rotor disc.

The comparison between the two different measuring units shows that there is fairly good
linearity between the readings. It signifies that both measurements give an accurate measure
of the shift in gap clearance. However, the determination of an exact zero-point was more
uncertain due to uncertainties associated to different conditions during pre-calibration and
running. The gap sensors were typically pre-calibrated in a pure water environment before
the pulp suspension was pumped through the refiner. The assembly of the sensors gave
additional challenges in order to fit them into pre-machined holes in the different plates.
Thus, initial offset values were observed prior to some of the trials. That is the reason for
the term offset removed at the y-axis in Figure 4. The estimated initial offset is removed.
Thermal effects like the expansion coefficients of the discs and the shaft assembly may
have given additional offset values. However, in the future it is believed that the advantage
of the automatic calibration of the gap sensors balances such effects especially if the
calibration is conducted both in advance and in the end of each trial. The offset associated
with the original gap displacement sensor is assumed to be more uncertain. The
investigation as shown in Figure 4.1 enhances this interpretation. Figure 4.1 shows that the
measures from the original gap displacement sensor varied more than the corresponding
data from the AGS devices. However, both measurements were more accurate, or with
other words the discrepancy in offset was less significant, when different trials were run
without opening (plate change) and restarting the refiner. The difference in offset values
was clearly revealed through the measurements of the gaps during two different runs with
different refiner plates. A smaller offset is revealed through the lines in pairs (Figure 4.1)
which corresponded to two different runs with only one refiner plate pattern.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

24 (41)

Figure 4.1. The sum of the values recorded by the gap sensors is compared with the
data from the disc position measuring device.

Besides the gap measurements, Figure 4.1 shows another result as well. This result strongly
indicates that the high-freeness pulp demanded a larger plate gap than the low-freeness
pulp. This was especially clear at the lowest specific edge load values i.e. when the gap was
large. Similar conclusions are made regarding the difference between softwood and
hardwood pulps (Nordman et al. 1981).

The next figures show that the Beloit DD refiner was not working ideally. The twin-
concept with two refining zones and a floating rotor disc gave not always similar refining
conditions in the two refining zones. Figure 4.2 shows that the two gaps were almost equal
at the lowest energy consumption whereas the subsequent progress at higher loads deviated
considerably. This signifies that the refining conditions differed between the zones at
higher loads. The same tendency is revealed in Figure 4.3.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

Figure 4.2. Plate clearance measurements versus the motor load given in the electric
current consumption denominated ampere.

Figure 4.3. Plate clearance measurements versus the motor load given in the electric
current consumption denominated ampere.

Figure 4.3 shows also that the initial gap clearance differed. It means that the rotor disc had
not the same position as observed during the previous run. The difference between these
runs was only that the pulps were of different quality. However, the low-freeness pulp was
STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

26 (41)

run after the high-freeness pulp without opening the refiner. The runs were only
distinguished by an ordinary shut-down including clean-up by pumping of water through
the refiner. Thus, it is reasonable to questioning the equality of the running conditions in
this refiner at every new start-up.

Another concern is revealed through the observations shown in Figures 4.4 and 4.5. The
first figure shows the gap measures from each of the refining zones. Despite of the
different gap clearances in the two refining zones, the slope of the linear regression curves
is almost equal. This signifies that the gap displacement responded to the same extent i.e.
the rotor disc was moving when the movable stator disc was forced to move. This was
observed through an increasing axial thrust. The opposite response was not tested, and it is
uncertain whether a hysteresis effect was prevailing. However, the next observation, as
shown in Figure 4.5, did not enhance the interpretation of a free floating rotor.

Figure 4.4. Plate clearance measurements versus the motor load given in the electric
current consumption denominated ampere.

Figure 4.5. Plate clearance measurements versus the motor load given in the electric
current consumption denominated ampere.

The slopes of the regression curves in Figure 4.5 were not equal which indicates that the
movement of the rotor was affected by other factors than the axial thrust alone. Different
amount of pulp in the two refining zones may have contributed to the inequality of the gap
According to Innventia Confidentiality Policy this report is public since 2011-02-04

displacement during influence of a movable stator disc. Thus, it can be assumed that the
double disc concept introduces an additional dimension regarding the uncertainty, and in
turn the homogeneity of the pulp quality can be questioned.

A different approach to use the gap measurements in the investigation of the refining
performance is shown in Figure 4.6. The measures of the gap clearance at constant specific
edge loads signify that the gap clearance was dependent of the flow rate at the lowest edge
loads, while it was independent of the flow rate at higher edge loads. However, this is not
enough information to make a categorical conclusion. Beneath the data points in this figure
more information is available. The motor load was almost constant at the lowest edge load,
while the motor load increased at increasing flow rate at the highest edge load. The specific
energy consumption was reduced in both cases, while the specific edge load theory
assumed constant values. Thus, the reliability of the specific edge load theory should be
questioned as a control tool at every running condition in the present refiner.

Figure 4.6. The recorded gap clearances during constant specific edge loads and
increasing flow rates show relationships which can question the reliability of
the SEL-theory at various process conditions.

5.1.2 Vibration measurements


Vibration measurements were accomplished in order to investigate different responses at
varying process conditions including different refiner plate patterns. The frequency spectra
are believed to show a fingerprint of the refining behaviour in the refining zone. Some
frequency peaks can directly be corresponded to the movement of the rotor bars at

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

28 (41)

different radial positions of the refiner plates. In addition, the overall vibration level gives
information of the stability of the process performance.

Figure 4.7 shows frequency information subtracted from recordings from different process
conditions which are related to the quality of the corresponding pulp samples. The
assessment of the relationship is that the best pulp quality, represented by the tensile index
of the handsheets, was obtained when the vibration level i.e. the height of the frequency
peaks, was relatively small. In addition, the refining performance seemed to give the best
pulp quality when the dominating peaks in the frequency spectra corresponded to the bar
movement of the bars located close to the periphery. Both of these factors are fulfilled in
the upper frequency spectrum of Figure 4.7. The lower spectrum indicates that the main
refining area occurred earlier in the refining zone since the dominating frequency peak
corresponds to a smaller number of rotor bars than what is the number of bars at the
periphery. However, the lower frequency spectrum shows also frequency peaks which can
be associated to the impact of fibrous material at the plate periphery. The latter may be
explained by the fact that the accelerometer did not separate vibrations generated from
both sides of the refiner. The vibrations from each side were probably not similar. The gap
clearances indicate that the refining performance was different in the two refining zones.

Figure 4.7. Vibrations of the refiner housing recorded at different process conditions
are related to the pulp quality of the corresponding pulp samples.

Table 4.1 shows supporting information about the actual frequencies. Both of the runs
were conducted at a measured rotational speed of 707 rpm (700 rpm in the DCS from
ABB Inc.). The frequency spectra show several different peaks which can be identified
based on the knowledge of the plate pattern. Additional attention can also be addressed to
According to Innventia Confidentiality Policy this report is public since 2011-02-04

the sideband frequencies which are assumed generated by flow disturbances from low-
frequency sources i.e. bolt holes, flow channels, breaker bars. The present recordings imply
that the radial directed open spaces from the inlet to the periphery also called flow channels
influence the flow of pulp through the refining zone frequently. Thus, their periodical
influence appears in the frequency spectra.

The broad band of frequencies appearing at somewhat higher frequencies than the
periphery frequency contains additional information. The broad band implies that the
source was of a stochastic nature. The origin of these frequencies has not been identified.

Table 4.1. Characteristic properties of the plates and their corresponding frequencies.

Medium bars: Ls=8.4 km/rev Fine bars: Ls=15.7 km/rev

Nos. bars Rotational speed, Nos. bars per sec.


Positions of Nos. of plates
per plate rotations per sec. (Frequency, Hz)
bars per revolution
M F 707 rpm M F
Pheriphery 64 84 3016.5 3959.2
At bolt holes 48 56 2262.4 2639.5
Inlet 32 44 1508.3 2073.9
Breaker bars 3 4 141.4 188.5
4 11.78
Other Nos. Nos. per sec.
characteristics per plate (Frequency, Hz)
Flow channels 4 188.5 188.5
Bolt holes 2 94.3 94.3

Figure 4.8 shows the variation of the signal collected from the accelerometer. The standard
deviation values from different recordings at various process conditions are shown in the
figure. The result shows that the vibrations increased with increasing load. In addition, the
plate pattern as well as the pulp quality i.e. low or high original freeness, affected the
vibration level. The refiner plates with the fine bars gave smaller vibrations than the plates
with medium size bars. The latter plates gave the highest vibrations when the low freeness

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

30 (41)

pulp was treated in the refiner. This seems to be connected to the smaller plate gap that
was prevailing during this run as shown in Figure 4.1.

Figure 4.8. The variation of the vibrations recorded from the refiner housing is
represented by the standard deviation values.

5.2 Trial 2
The main focus of the second trial was connected to the test of fibre-optic pressure sensors
in the refining zone as reported by Eriksen (2005). However, the high-speed data
acquisition system was used to collect additional data too. The gap clearance sensors as well
as vibrations of the refiner housing were recorded simultaneously.

5.2.1 Gap measurements – DCS data


Recordings from the gap sensors and the original gap displacement sensor are compared in
Figure 4.9. The data is captured entirely from the distributed control system (DCS) of the
pilot plant. It is evidently that the total offset was largest for the gap sensors. However, as
discussed previously, the initial offset of the gap sensors contributed to approximately
1 mm in this trial. This offset is not estimated more exactly and it is not removed in the
present data analyses. The main objective with Figure 4.9 is to show the influence of the
rotational speed and the flow rate on the gap clearance. Measures were captured at three
different specific edge loads (0.5, 0.7 and 0.9 Ws/m) in each refining series. The prevailing
refiner plate pattern consisted of coarse bars with dams.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

Figure 4.9. The sum of the values recorded by the gap sensors is compared with the
data from the disc position measuring device.

It seems to be rather clear that a higher rotational speed caused the gap clearance to
decrease assuming that the flow rate was constant. On the other hand, the flow rate
influences the gap clearance too. Figure 4.9 shows also that the gap clearance increased
when the flow rate increased. The latter assumes that the rotational speed was remained
constant.

5.2.2 Gap and vibration measurements


Figure 4.10 shows both vibration and gap clearance measures, and the objective is to
illustrate the difference obtained by changing the rotational speed of the refiner. Increased
speed of the refiner gave significantly higher vibrations. The term 2 x gap on the x-axis
represents the sum of the two gap clearance measures. From the figure it is also revealed
that the gap clearance at equal specific edge load increased when the rotational speed was
increased. The present plate pattern consisted of a high number of fine bars with an edge
length larger than 18 km per revolution.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

32 (41)

Figure 4.10. The variation of the vibrations recorded from the refiner housing is
represented by the standard deviation values. The vibrations are related to
the sum of the two gap clearance measures.

Figure 4.11. The variation of the vibrations in the refiner housing is related to the sum
of the two gap clearance measures.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

The other plate pattern that was run in this trial consisted of coarse bars with a relative
high number of full surface dams. The corresponding plot of vibrations against the sum of
the gap clearance measures is shown in Figure 4.11.

A comparison between the results in Figures 4.10 and 4.11 shows that the plates with the
coarse bars incorporated dams gave a much higher vibration level. At higher loads and a
high rotational speed the vibrations increased to a substantial level far higher than observed
in the previous trials. The dams probably obstructed the pulp flow through the grooves
such that a considerably mixing occurred in the refining zone. In addition, a lot more of the
pulp had to be transported in the plate gap compared to the amount of pulp present in the
gap when other refiner plates without dams were used. Thus, as in addition to the high
rotational speed, the vibrations of the refiner housing increased drastically.

5.2.3 Vibration frequencies


Figure 4.12 shows typical frequency spectra from the recordings of vibrations when the
refiner plates with the fine bars were used. The difference between the upper and lower
spectrum is more or less determined by the pertinent rotational speed.

Figure 4.12. Vibrations of the refiner housing represented by the frequency spectra
recorded at different rotational speed.

Both spectra in the figure show that the frequencies associated to the movement of the
bars located nearby the periphery of the plates dominated. The different frequencies were
solely connected to the different rotational speed of the rotor disc. Table 4.2 is made in
order to support the comprehension of the frequency readings.
STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

34 (41)

Table 4.2. Characteristic properties of the plate and their corresponding frequencies.

Rotational speed, Nos. bars per sec.


Nos. of
Positions of Nos. bars rotations per sec. (Frequency, Hz)
plates per
bars per plate 707 900
revolution 707 rpm 900 rpm
rpm rpm
Periphery 88 4147.7 5280
At bolt holes 64 3020.0 3840
Inlet 44 2073.9 2640
Breaker bars 4 188.5 240
4 11.78 15.0
Other Nos. Nos. per sec.
characteristics per plate (Frequency, Hz)
Flow channels 4 188.5 240
Bolt holes 2 94.3 120

5.3 Trial 3
This trial was also mainly addressed to investigations of the fibre-optic pressure sensors.
However, additional information was search for in order to learn more of the process
behaviour. A third refiner plate pattern from J&L Fiber Services was applied this time. The
captured data is connected to four different flow rates each refined at three different loads
(50, 70, 90 kW net) at constant rotational speed. The original pulp was a high-consistency
RTS pulp.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

5.3.1 Gap measurements


The sum of the gap clearance measures and the recording of the original gap displacement
are shown in Figure 4.13. Previously concerns associated to initial offsets due to the
assembly of the sensors were surmounted before this trial. Thus, the difference obtained
between the two measurements indicates that the offset in one or both of the
measurements was approximately 0.5 mm. Most of this offset is interpreted to be
connected to the original gap displacement sensor. Compared to the recordings in
Figures 4.1 and 4.9, the main difference, besides the initial offset values, was connected to
the gradient of the regression lines. The present gradient differed considerably from the
previous observations as well as from the expectation of an ascent equal to one. The origin
of this discrepancy is not found. However, a nonlinearity relationship was demonstrated as
shown in Figure 3.11. The differences between different trials may also be objected to the
fact that the gap sensors were prototypes and the trials were used to develop their
performance too.

Figure 4.13. The sum of the values recorded by the gap sensors is compared with the
data from the disc position measuring device.

5.3.2 Gap and vibration measurements


Figure 4.14 shows the relationship between the sum of gap clearance measures and the
overall vibration level represented with the standard deviation of the recordings from the
accelerometer. The adaptation of the fitted curve to all of the data points gives an optimum
at gap clearances of approximately 200 µm i.e. the sum of gap clearances is approximately
400 µm. Based on the pulp quality analyses, reported by Hammar (2005), the highest tensile
indexes were reached by the pulps refined at the corresponding refining conditions (0.5 and
0.7 Ws/m at 200 l/min and 0.5, 0.7 and 0.9 Ws/m at 300 l/min). The latter pulps are
STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

36 (41)

associated with the three open triangles in Figure 4.14. When the flow rate was increased, it
is assumed that the load was not high enough and the gap clearance was too small to
maintain a sufficient fibre development. Thus, it is conceivable that the gap clearance
should reach an adequate level regardless of the flow rate. It is still hypothetical if this level
can be associated with the optimum of the gap and vibration relationship as indicated in
Figure 4.14.

Figure 4.14. The variation of the vibrations recorded from the refiner housing is
represented by the standard deviation values. Here the vibrations are related
to the sum of the two gap clearance measures.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

6 Summary
The present study has revealed some features of the Beloit DD refiner as well as the
different refiner plates investigated. This is made possible through the gap clearances and
the vibration measurements. Among the most important it is shown that the twin-concept
with two refining zones and a floating rotor disc gave not always similar refining conditions
in the two refining zones. It is reasonable to questioning the equality of the running
conditions in this refiner due to an observed asymmetrical position of the rotor disc. In
addition, it is strongly indicated that the movement of the rotor disc was uneven in some
runs. This signifies that the rotor movement was affected by other factors than the axial
thrust alone. The reason for the inequality of the gap displacement may be caused by
different amount of pulp in the two refining zones. If so, the refining energy may be
distributed more evenly by the different gap clearances in the two refining zones. However,
it remains to be demonstrated that the uneven gap clearances are compensated by different
amount of pulp in the two refining zones. It is reasonable to question the homogeneity of
the double disc concept and evaluate whether an additional dimension regarding the
uncertainty is introduces compared to single disc operation. The reliability of the specific
edge load theory should also be questioned as a control tool at every running condition in
the present refiner.

Despite that the modified TDC sensors were objected to development during the
accomplishment period of the trials, it can be concluded that a considerable advantage of
the AGS-sensor is that the zero-point setting can be checked and adjusted when the refiner
is running. The uncertainty in zero-point setting is much less for the AGS system
compared to what can be achieved with an external position sensor i.e. original gap
displacement sensor. However, the impression is that both measurements give an accurate
measure of the shift in gap clearance i.e. the measures of the dynamic behaviour of the disc
movement seem to be trustworthy.
Among other results, it is shown that a higher rotational speed caused the gap clearance to
decrease at a constant flow rate. On the other hand, the gap clearance increased when the
rotational speed was increased if the specific edge load was kept constant. The flow rate
influenced the gap clearance too. The gap clearance increased when the flow rate increased.
The latter was observed at similar rotational speed of the refiner. Another observation
strongly indicated that high-freeness pulp demanded a larger plate gap than low-freeness
pulp.

Vibration measurements were accomplished in order to investigate different responses at


varying process conditions including different refiner plate patterns. The frequency spectra
are believed to show a fingerprint of the refining behaviour in the refining zone. Some
frequency peaks correspond directly to the movement of the rotor bars at different radial
positions of the refiner plates. The frequency spectra showed several different peaks which

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

38 (41)

were identified based on the knowledge of the plate pattern. A main conclusion was that
the frequencies associated to the movement of the bars located nearby the periphery of the
plates dominated. In addition, sideband frequencies were regularly observed. Sideband
frequencies are assumed generated by flow disturbances from low-frequency sources such
as bolt holes, flow channels and breaker bars. The present recordings imply that the radial
directed open spaces from the inlet to the periphery also called flow channels influenced
the flow of pulp through the refining zone. Their periodical influence appeared commonly
in the frequency spectra.
The refiner plates with the highest bar angle seemed to generate a different frequency
pattern than the others. The periphery frequencies dominated here as well. However, a
broad band of frequencies appeared additionally at somewhat higher frequencies than the
periphery frequency. The broad band implies that the source was of a stochastic nature, but
the origin of these frequencies has not been identified.
Another result showed that increased speed of the refiner gave significantly higher
vibrations. The plates with the coarse bars incorporated dams gave the highest vibration
level. At higher loads and a high rotational speed the vibrations increased to a substantial
level far higher than observed in the previous trials. The dams probably obstructed the
pulp flow through the grooves such that a considerably mixing occurred in the refining
zone. In addition, a lot more of the pulp had to be transported in the plate gap compared
to the amount of pulp present in the gap when other refiner plates without dams were
used.

The results indicate that a relationship between the sum of gap clearance measures and the
overall vibration level was present. The vibration level reached an optimum at a gap
clearance of approximately 200 µm i.e. the sum of gap clearances was approximately 400
µm. It is conceivable that the gap clearance should reach an adequate level regardless of the
flow rate. It is still hypothetical if this level can be associated with the optimum of the gap
and vibration relationship.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

7 References
Dahlqvist, G. and Ferrari, B. (1981): “Mill operating experience with a TMP refiner control
system based on a true disc clearance measurement”, International Mechanical Pulping
Conference, pp. 1-14.
Eriksen, O., Gregersen, Ø. and Krogstad, P.-Å. (2006): “Pressure and vibration in the
refining zone of a TMP refiner – influence of the fibre flow”, Nord. Pulp Paper
Res. J., 21(1):90-98.
Eriksen, O. (2005): “Mechanisms in the refining zone for development of physical
properties of TMP fibres in a low-consistency refiner. Pressure measurements in the
refining zone”, Technical report STFI-Packforsk no. 114, 28 p.
Eriksen, O. and Hammar, L.-Å. (2005): “Literature survey. Measurement techniques
suitable for the refining zone of disc and conical LC refiners”, Technical report STFI-
Packforsk no. 113, 39 p.
Eriksen, O. (2003): “High-frequency pressure measurements in the refining zone of a high-
consistency refiner”, Doctoral thesis, Department of Energy and Process Technology,
Norwegian University of Science and Technology, 50(04), 303 p.,
ISBN: 82-471-5601-6.
Hammar, L.-Å. (2004): “LC refining mechanisms – Literature review”, STFI report MEC
No. 42, 21 p.
Hammar, L.-Å. (2005): “Mechanisms in the refining zone for development of physical
properties of TMP fibres in a low-consistency refiner. Plate pattern, plate clearance
and pulp quality”, Technical report STFI-Packforsk no. 108, 29 p.
Jackson, M., Falk, B. and Danielsson, O. (1986): “Refining systems, pulp quality can be
tailored to specific end-use needs”, Pulp and Paper, 60 (6):60-63.
Mohlin, U.-B. (2006a): “LC-refining of chemical pulps – refining intensity and gap
clearance”, Industrikontakt, STFI-Packforsk, Stockholm, Sweden, No. 1, pp. 9-11.
Mohlin, U.-B. (2006b): “Refining intensity and gap clearance”, 9th International Refining
Conference, Pira.
Nordman, L., Levlin, J.-E., Markonen, T. and Jokisalo, H. (1981): “Conditions in an LC-
refiner as observed by physical measurements”, Pap. Puu, 63 (4):169-180.
Pettersen, G.W. (1986): “Noise emanation from conical refiners”, Pulp Paper Can.,
87(9):T336-T340.
Pettersen, G.W. and Gunstrøm, A. (1980): “Maleorganer og støy”, SCAN report, No. 235,
PFI, Oslo, Norway, 57 p.
Stationwala, M.I., Atack, D., Wood, J.R., Wild, D.J. and Karnis, A. (1979): “The effect of
control variables on refining zone conditions and pulp properties”, International
Mechanical Pulping Conference, pp. 93-109.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a LC refiner.
Gap clearance and vibration measurements.

40 (41)

Strand, B.C.and Mokvist, A. (1987): “Control and optimization of conical disc refiners”,
International Mechanical Pulping Conference, pp.11-18.
Strand, B.C. and Hartler, N. (1985): “Modelling and optimization of full scale chip
refining”, International Mechanical Pulping Conference, pp.46-54.
According to Innventia Confidentiality Policy this report is public since 2011-02-04

Appendix A:
STFI-Packforsk EuroFEX Laboratory Report. Trial 1 – Week 36, 2004
Svensson, B. (2004a): Sammanställning av data. LC-malning av TMP i Beloit DD, vecka 36,
2004. STFI-Packforsk EuroFEX Laboratory Report, 8 p.

Appendix B:
STFI-Packforsk EuroFEX Laboratory Report. Trial 2 – Week 41, 2004
Svensson, B. (2004b): Sammanställning av data. LC-malning av TMP i Beloit DD, vecka
41, 2004. STFI-Packforsk EuroFEX Laboratory Report, 8 p.

Appendix C:
STFI-Packforsk EuroFEX Laboratory Report. Trial 3 – Week 22, 2005
Svensson, B. (2005): Sammanställning av data. LC-malning av TMP och RTS i Beloit DD,
vecka 22, 2005. STFI-Packforsk EuroFEX Laboratory Report, 8 p.

STFI-Packforsk AB
Visit Drottning Kristinas väg 61 | Mail Box 5604, 114 86 Stockholm, Sweden
Tel +46 8676 70 00 | Fax +46 8411 55 18 | www.stfi-packforsk.se | info@stfi.se
VAT no. SE5566603110901
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt

EuroFEX

Sammanställning av data

LC-malning av TMP i Beloit DD, vecka 36, 2004

Kund: Lars-Åke Hammar

Projektledare: Lars-Åke Hammar

Projektnummer: 286 08 20

Uppdragsdatum: Vecka 36, 2004

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00

H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc


According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Resultatsammanställning

LC-malning av TMP i Beloit DD, vecka 36, 2004

Innehållsförteckning
Sid

1 Sammanfattning 3

2 Försöksbeskrivning 4
2.1 Syfte 4
2.2 Försöksupplägg 4
2.3 Maskinkonfiguration 4
2.4 Råvaror 5
2.5 Provtagning & analyser 5
2.6 On-line mätning 5

3 Försöksdata 6

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00 2


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

1 Sammanfattning
Passagemalning vid 3,5 % massakoncentration och hög temperatur (ca.85°C) i Beloit
DD-kvarn av en högfreeness och en lågfreeness TMP från Braviken.

Syftet var att studera effekten av olika malsegment och specifik energiinsats vid
givna kanbelastningar. De specifika energiinsatserna erhölls genom variation av
produktionsflödet under konstant effekt i kvarnen vid olika kantbelastning.

På grund av olikheter i tomgångsenergi beroende på segmenttyp och massa, blev de


planerade börvärdesnivåerna för specifik energiinsats (250-200-150 kWh/t br.) något
förskjutna varför ärvärdesnivån blev högre, vilket framgår av försökstabellerna 1-4.

Under respektive försök loggades kontinuerligt erforderliga maskin och processdata


samtidigt som en triggad markering gjordes i loggfilerna vid varje provuttag vilket
underlättar identifieringen av respektive prov vid utvärdering.

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00 3


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

2 Försöksbeskrivning

2.1 Syfte
Studera effekten av olika segment och specifik energiinsats vid givna
kantbelastningar på en hög- respektive lågfreeness TMP.

2.2 Försöksupplägg
Varmvatten från massalagret, 55-60°C, lagrades i A-karet för uppvärmning via
värmeväxlare till 90-95 °C.
Massa från storsäckar (ca.44% torrhalt) matades med bandtransportör till B-karet
varefter varmvatten, ca.95°C, från A-karet tillsattes och uppslagningen startade.
Efter uppslagning och koncentrationsbestämning kördes under fyra likartade
försöksomgångar, malserier med varierande specifik energiinsats i tre steg vid olika
kantbelastning på två olika segmenuppsättningar. Ett mediumfint segment av typ:
24101/102 med kantlängden 8,38 km/rev samt ett fint segment av typ: 24113/114
med kantlängden 15,72 km/rev.

• Massa: Förstastegs TMP högfreeness (S1) samt andrastegs TMP lågfreeness


(S2) från Braviken
• Kvarn: Beloit DD
• Segment: 24101/102 BJ (medium) & 24113/114 BJ (fine)
• Energiinsats: Börvärde 150 - 200 - 250 kWh/t br.
• Kantbelastning: 0,3 - 0,5 - 0,7 - 0,9 - 1,1 Ws/m
• Mäldkoncentration: 3,5 %
• Varvtal: 700 rpm
• Temperatur: 80-90°C

2.3 Maskinkonfiguration

A-kar B-kar

Värmeväxlare Beloit DD

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00 4


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

Förvärmt vatten 55-60°C från massalagret tillfördes A-karet och värmdes via
värmeväxlare i cirkulation till 90-95°C för spädning och uppslagning av massa i B-
karet. Malningen genomfördes som passagemalning från B-kar till A-kar. Efter varje
avslutad försöksserie dumpades massan från A-kar till E-kar.

2.4 Råvaror
Förstastegs högfreeness och andrastegs lågfreeness TMP från Braviken.

2.5 Provtagning & analyser


Innan start av varje försök (1-4) uttag av 0-prov från B-karet. Därefter 2 x 5 liter på
varje flödesnivå i respektive malserie. Analyser som utfördes på utvalda prov var
massa-koncentration, varmuppslagning och freenessmätning.

2.6 On-line mätning


Erforderliga kvarn- och processdata loggades kontinuerligt under varje försök. Vid
varje provuttag markerades respektive provperiod i en speciell trigg-fil för att lättare
kunna identifieras vid beräkning och studier av egenskapsförändringar.

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00 5


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

3 Försöksdata

Försök 1. Kördag 04-08-31


Serie, Massa Segment Kant- Tom- Belastn. Prod. Flöde, SE,
No: typ belastning, gång, kW br. t/h l/min kWh/t
Ws/m kW br.
20 TMP (S1) 24101/102 70 -------- 0,336 160 0-prov
21 0,5 70 109 0,336 160 324
22 “ “ “ 0,420 200 260
23 “ “ “ 0,560 267 195

31 TMP (S1) 24101/102 0,7 70 128 0,412 196 310


32 “ “ “ 0,515 245 248
33 “ “ “ 0,687 327 186

41 TMP (S1) 24101/102 0,9 70 148 0,492 234 301


42 “ “ “ 0,615 293 241
43 “ “ “ 0,820 390 180

51 TMP (S1) 24101/102 1,1 70 168 0,572 272 294


52 “ “ “ 0,715 341 235
53 “ “ “ 0,953 454 176

Försök 2. Kördag 04-09-01


Serie, Massa Segment Kant- Tom- Belastn. Prod. Flöde, SE,
No: typ belastning, gång, kW br. t/h l/min kWh/t
Ws/m kW br.
70 TMP (S2) 24101/102 63 ---- 0,336 160 0-prov
71 0,5 63 109 0,336 160 324
72 “ “ “ 0,420 200 260
73 “ “ “ 0,560 267 195

81 TMP (S2) 24101/102 0,7 63 128 0,412 196 310


82 “ “ “ 0,515 245 248
83 “ “ “ 0,687 327 186

91 TMP (S2) 24101/102 0,9 63 148 0,492 234 301


92 “ “ “ 0,615 293 241
93 “ “ “ 0,820 390 180

101 TMP (S2) 24101/102 1,1 63 168 0572 272 294


102 “ “ “ 0,715 341 235
103 “ “ “ 0,953 454 176

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00 6


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

Försök 3. Kördag 04-09-02


Serie, Massa Segment Kant- Tom- Belastn. Prod. Flöde, SE,
No: typ belastning, gång, kW br. t/h l/min kWh/t
Ws/m kW br.
121 TMP (S1) 24113/114 0,5 57 149 0,488 232 305
122 “ “ “ 0,610 291 244
123 “ “ “ 0,813 387 183

131 TMP (S1) 24113/114 0,7 57 185 0,632 301 292


132 “ “ “ 0,790 376 234
133 “ “ “ 1,053 501 176

141 TMP (S1) 24113/114 0,9 57 202 0,780 371 285


142 ” ” ” 0,975 464 207

Försök 4. Kördag 04-09-03

Serie Massa Segment Kant- Tom- Belastn. Prod. Flöde SE,


, typ belastning gång, kW br. t/h , kWh/t
No: , kW l/min br.
Ws/m
150 TMP (S2) 24113/114 41 -------- 0,340 162 0-prov
153 0,3 41 96 0,567 270 169

161 TMP (S2) 24113/114 0,5 41 133 0,488 232 273


162 “ “ “ 0,610 291 218
163 “ “ “ 0,813 387 164

171 TMP (S2) 24113/114 0,7 41 169 0,632 301 267


172 “ “ “ 0,790 376 214
173 “ “ “ 1,053 501 160

181 TMP (S2) 24113/114 0,9 41 206 0,780 371 264


182 “ “ “ 0,975 464 211
183 “ “ “ 1,300 619 158

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00 7


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt

Vänligen returnera denna sida till STFI

Genom att denna sida returneras till EuroFEX kan vi försäkra oss om att
denna slutgiltiga sammanställning av data mottagits av Er.

Till: Börje Svensson

Från: Lars-Åke Hammar

Fax: +46 8 676 71 80

Här med intygas att Lars-Åke Hammar mottagit en slutgiltig


sammanställning av data från uppdrag vid STFI vecka 436.
Sammanställningen har levererats tillsammans med tillgänglig data, allt
enligt överenskommelse.

Lars-Åke Hammar

Skapat av Börje Svensson Skapat den 2004-09-10 13:23:00

H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam.ställ.rapp Beloit v436 L.ÅH.doc


According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt

EuroFEX

Sammanställning av data

LC-malning av TMP i Beloit DD, vecka 41, 2004

Kund: Lars-Åke Hammar

Projektledare: Lars-Åke Hammar

Projektnummer: 286 08 60

Uppdragsdatum: Vecka 41, 2004

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00

H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc


According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Resultatsammanställning

Innehållsförteckning
Sid

1 Sammanfattning 3

2 Försöksbeskrivning 4
2.1 Syfte 4
2.2 Försöksupplägg 4
2.3 Maskinkonfiguration 5
2.4 Råvaror 5
2.5 Analyser provtagning 5
2.6 On-line mätning 5

3 Kördata 6

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 2


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

1 Sammanfattning
Passagemalning vid 3,5 % massakoncentration och hög temperatur (ca.85°C) i Beloit DD-
kvarn av en 1á-stegs högfreeness TMP från Holmen-Braviken.
Syftet var att i två olika försök studera inverkan av varierat flöde vid konstant kant-
belastning och varvtal, samt inverkan av olika varvtal och kantbelastning vid konstant
flöde.
På grund av massabrist begränsades försöksomgång 1 till endast två serier, 10 och 20,
samt i försöksomgång 2 till tre serier, 50 t.o.m 70.
Under respektive försök loggades kontinuerligt erforderliga maskin och processdata
samtidigt som en triggad markering gjordes i loggfilerna vid varje provuttag vilket
underlättar identifieringen av respektive prov vid utvärdering.

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 3


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

2 Försöksbeskrivning

2.1 Syfte
Syftet var att i två olika försök på en 1á-stegs högfreeness TMP från Holmen-Braviken
uttagen under vecka 440, studera inverkan av varierat flöde vid konstant kantbelastning
och varvtal, samt inverkan av olika varvtal och kantbelastning vid konstant flöde.

2.2 Försöksupplägg
Varmvatten från massalagret (55-65 °C) lagrades i A-karet för ytterligare uppvärmning till
85-90 °C via värmeväxlare. Massa matades med transportör till B-karet varefter varm-
vatten från A-karet tillfördes och uppslagning startade. All malning gjordes därefter med
passage från B-kar till A-kar.
Ett referens nollprov togs ur B-karet efter varje uppslagning samt tre provuttag i varje
malserie. Efter uppslagning och konc.bestämning kördes i försöksomgång 1 på ett fint
segmentmönster (24EJ 113/114), två malserier med varierat flöde vid konstant
kantbelastning och varvtal.
Under försöksomgång 2, med grovt segmentmönster (24EJ 111/112), gjordes tre malserier
med varierad kantbelastning vid två konstanta varvtal och flödesnivåer.
• Massa: Förstastegs TMP från Holmen-Braviken
• Kvarn: Beloit DD
• Segment: 24EJ 113/114 (Fine), 24EJ 111/112 (Coarse-dams)
• Kantbelastning: 0,3-0,5-0,7-0,9 Ws/m
• Flöde försöksomgång 1: 900, 600, 300 l/min
• Flöde försöksomgång 2: 400 & 800 l/min
• Varvtal: 700 & 900 rpm.
• Massakoncentration: 3,6 % (omgång 1) & 3,7 % (omgång 2)
• Temperatur: ca. 85°C

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 4


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

2.3 Maskinkonfiguration

A-kar B-kar

Beloit DD
Värmeväxlare

2.4 Råvaror
Högfreeness 1á-stegs TMP från Holmen-Braviken uttagen i storsäck under
vecka 440.

2.5 Analyser provtagning


Efter varje uppslagning uttag av referens 0-prov från B-karet före malning. Därefter 2 x 5
liter på varje provpunkt i respektive malserie. Analyser som utfördes på utvalda prov var
massakoncentration, varmuppslagning och freenessmätning.

2.6 On-line mätning


Erforderliga kvarn- och processdata loggades kontinuerligt under varje försök. Vid varje
provuttag markerades respektive provperiod i en speciell trigg-fil för att lättare kunna
identifieras vid beräkning och studier av egenskapsförändringar.

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 5


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

3 Kördata
Försöksomgång 1: Segment 24EJ 113/114 (Fine)
Serie Kantbelastn. Flöde Varvtal Belastning Belastning Spec.energi
No: Bs, Ws/m l/min rpm kW, netto kW brutto kWh/t, netto
11 0,3 900 700 65 107 33
12 0,3 600 700 65 107 50
13 0,3 300 700 65 107 100

21 0,5 900 700 108 150 56


22 0,5 600 700 108 150 83
23 0,5 300 700 108 150 166
På grund av massabrist utgick provserie 30 & 40.

Försöksomgång 2: Segment 24EJ 111/112 (Coarse-dams)


Serie Kantbelastn. Flöde Varvtal Belastning Belastning Spec.energi
No: Bs, Ws/m l/min rpm kW, netto kW brutto kWh/t, netto
51 0,5 400 700 36 89 40
52 0,7 400 700 50 103 56
53 0,9 400 700 64 117 72

61 0,5 400 900 46 162 52


62 0,7 400 900 64 180 72
63 0,9 400 900 83 199 93

71 0,5 800 900 46 162 26


72 0,7 800 900 64 180 36
73 0,9 800 900 83 199 47
På grund av massabrist utgick provserie 80.

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 6


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 7


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00 8


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt

Vänligen returnera denna sida till STFI

Genom att denna sida returneras till STFI kan vi försäkra oss om att denna
slutgiltiga sammanställning av data mottagits av Er.

Till: Börje Svensson

Från: Lars-Åke Hammar

Fax: +46 8 676 71 80

Här med intygas att Lars-Åke Hammar mottagit en slutgiltig


sammanställning av data från uppdrag vid STFI vecka 441.
Sammanställningen har levererats tillsammans med tillgänglig data, allt
enligt överenskommelse.

Kund

Skapat av Börje Svensson Skapat den 2004-10-27 12:15:00

H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Sam. ställ rapp Beloit v441 LÅH.doc


According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt

EuroFEX

Sammanställning av data

LC-malning av TMP och RTS i Beloit DD,


vecka 22, 2005

Kund / Proj.ledare: Lars-Åke Hammar

Försöksledareledare: Börje svensson

Projektnummer: 286-12-60

Uppdragsdatum: 2005-05-30

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00

H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Resultatsammanställning

Innehållsförteckning
Sid

1 Sammanfattning 3

2 Försöksbeskrivning 4
2.1 Syfte 4
2.2 Försöksupplägg 4
2.3 Maskinkonfiguration 6
2.4 Råvaror 6
2.5 Analyser provtagning 6
2.6 On-line mätning 6

3 Kördata 7

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00 2


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

1 Sammanfattning
Passagemalning av två mekaniska massor, TMP S1 från Braviken och RTS från Hallsta,
i Beliot DD vid hög temperatur, ca. 85°C, och massakoncentrationen ca. 3,5 %.

Syftet var att studera inverkan av olika kantbelastning (Bs) i Beloit DD vid givna flöden,
varierande mellan 150 till 600 l/min. Under försöket gjordes samtidigt malspaltsmätning
enligt DAMETRIC, samt tryck- och pulsmätning med hjälp av EFPI-sensorer monterade i
ett av de fasta statorsegmenten.

På grund av extremt lågt flöde, 150 l/min, framgick att belastning av kvarnen till den
högsta kantbelastningen i försöksserie FPM 44-10 orsakade segmentkontakt.
Vid malning med RTS-massa har därför den flödesnivån (serie FPM 44-60) utgått ur
försöksserien.

Under respektive försök loggades kontinuerligt erforderliga maskin- och processdata


samtidigt som en triggmarkering gjordes i loggfilerna för att lättare identifiera prov-
perioden vid utvärdering.

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00 3


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

2 Försöksbeskrivning

2.1 Syfte
Studera inverkan av olika kantbelastning (Bs) i Beloit DD vid givna flöden, varierande
mellan 150 till 600 l/min. Under försöket gjordes samtidigt malspaltsmätning enligt
DAMETRIC, samt tryck- och pulsmätning med hjälp av EFPI-sensorer monterade i ett av
de fasta statorsegmenten.

2.2 Försöksupplägg
Varmvatten från massalagret (55-60 °C) lagrades i A-karet för ytterligare uppvärmning till
95°C via värmeväxlare.
Massan matades med bandtransportör till B-karet varefter varmvatten från A-karet till-
fördes och uppslagningen startade. Massans kylande effekt gjorde att malningarna
utfördes vid en lägre temperatur, ca.85-87°C.
All malning gjordes med passage från B-kar till A-kar. Energiinsatsen under försöken
baserades på en teoretisk koncentration av 35g/l, vilket vid senare utvärdering korrigerats
med den uppmätta massakoncentrationen.

Referens nollprov togs efter varje uppslagning ur B-karet samt efter malningens avslut.
På varje flödesnivå gjordes ett provuttag (2*5liter) vid respektive energiinsats.

• Massa: Förstastegs högfreeness, S1 från Braviken och RTS från Hallsta


• Kvarn: Beloit DD
• Segment: 24EJ 145 C40 / 24EJ 146 C40 Ls: 8,539 km/rev
• Flöde: ( 150 ) – 200 – 300 – 400 – 600 l/min
• Kantbelastning ( Bs ) : Börv. 0,5 – 0,7 – 0,9 Ws/m
• Massakoncentration: Teoretisk 35 g/l
• Varvtal: 700 rpm
• Temperatur: 85-87°C vid uppslagning ( 95°C på vatten före uppslagning)

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00 4


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

Försöksmatris avseende båda massatyperna.


TMP S1 (Braviken) serie 10 - 50, RTS ( Hallsta ) serie 70 - 100

Serie Flöde Prod. SEL Effekt SE Anm:


No: l/min t/h Ws/m kW ne. kWh/t ne.
0-prov ur B-kar före körning
10 & (60) 150 0,315 0,5 50 159 Serie 60 utgick p.g.a lågt flöde.
0,7 70 222 Omöjligt belasta kvarn p.g.a
0,9 90 286 segmentkontakt

20 & 70 200 0,420 0,5 50 119


0,7 70 167
0,9 90 214

30 & 80 300 0,630 0,5 50 79


0,7 70 111
0,9 90 143

40 & 90 600 1,260 0,5 50 40


0,7 70 56
0,9 90 71

50 & 100 400 0,840 0,5 50 59


0,7 70 83
0,9 90 107
0-prov ur B-kar efter körning

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00 5


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

2.3 Maskinkonfiguration

A-kar B-kar

Beloit DD

Värmeväxlare

2.4 Råvaror
Högfreeness TMP S1 från Braviken och RTS från Hallsta uttagna i storsäck under vecka
520 och 521.

2.5 Analyser provtagning


Efter varje uppslagning uttag av referens 0-prov från B-karet före och efter malningens
avslut. På varje provpunkt i respektive malserie uttag av 2*5 liter.

2.6 On-line mätning


Erforderliga kvar- och processdata loggades kontinuerligt under varje försök.
Under respektive försök loggades kontinuerligt erforderliga maskin- och processdata
samtidigt som en triggmarkering gjordes i loggfilerna för att lättare identifiera prov-
perioden vid utvärdering.

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00 6


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt
Sammanställning av data

3 Kördata

Körpunkt Ws / m kW br. Flöde, l/min Prod, t/h kW tomg. kWh/t netto

TMP (S1)

FPm 44-10 0,5 120,8108 150 0,279 71 179


-11 0,69 139,739 150 0,279 71 246
-12 0,97 167,633 150 0,279 71 346

FPm 44-20 0,48 118,8184 200 0,372 71 129


-21 0,72 142,7276 200 0,372 71 193
-22 0,89 159,6633 200 0,372 71 238

FPm 44-30 0,5 120,8108 300 0,558 71 89


-31 0,7 140,7352 300 0,558 71 125
-32 0,9 160,6595 300 0,558 71 161

FPm 44-40 0,53 123,7995 600 1,116 71 47


-41 0,7 140,7352 600 1,116 71 62
-42 0,89 159,6633 600 1,116 71 79

FPm 44-50 0,52 122,8033 400 0,744 71 70


-51 0,71 141,7314 400 0,744 71 95
-52 0,9 160,6595 400 0,744 71 121

TMP (RTS)

FPm 44-70 0,49 126,8146 200 0,436 78 112


-71 0,71 148,7314 200 0,436 78 162
-72 0,92 169,6519 200 0,436 78 210

FPm 44-80 0,5 127,8108 300 0,653 78 76


-81 0,72 149,7276 300 0,653 78 110
-82 0,91 168,6557 300 0,653 78 139

FPm 44-90 0,52 129,8033 600 1,307 78 40


-91 0,75 152,7163 600 1,307 78 57
-92 0,9 167,6595 600 1,307 78 69

FPm 44-100 0,5 127,8108 400 0,871 78 57


-101 0,71 148,7314 400 0,871 78 81
-102 0,92 169,6519 400 0,871 78 105

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00 7


H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
Konfidentiellt

Skapat av Börje Svensson Skapat den 2006-12-08 10:25:00

H:\Arkiv\Rapporter\STFI-Packforsk\STFI_115_05\Samställn.rapp v522.doc
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to Innventia Confidentiality Policy this report is public since 2011-02-04
According to STFI-Packforsk's Confidentiality Policy this report is assigned category

Mechanisms in the refining zone for development of physical properties of TMP fibres in a low-consistency
refiner

42 (42)

STFI-Packforsk Database information


Title
Mechanisms in the refining zone for development of physical properties of TMP fibres in a
low-consistency refiner. Gap clearance and vibration measurements.

Author
Oddbjørn Eriksen, Bertil Roos

Abstract

Keywords
Refining, low-consistency, gap clearance, vibrations, frequency analysis
Classification

Type of publication

Report number
115

Publication year
12 2005

Language
English

File name: H:\Arkiv\Rapporter\STFI-


Packforsk\STFI_115_05\LC_RefMec_Vib_Gap_mall_rev0.doc
Updated: 2007-01-04 13:20
Printed: 2007-01-04 13:20

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy