CA2108321A1 - Mechanical pulp fractionation with a spray disk filter - Google Patents

Abstract

Abstract A method for removing coarse fibers from mechanical pulp comprising the steps of spraying the pulp containing coarse fibers under pressure against a rotating disk filter such that a first fraction of the pulp passes through the filter and constitutes an accepts fraction and a second fraction of said pulp is retained on the filter and constitutes an enriched concentration of the coarse fibers and constitutes a rejects fraction, and removing the accepts fraction and the rejects fraction from the filter.

Description

:

PATENT
MECEIANICAL PULP FRACTIONATION WIT~
A SPRAY DISK FILTER

Background of the Invention This invention generally relates to wood pulp and, more particularly, to a process for removing the coarse fraction from a ~echanical pulp.

Mechanical pulp often contains significant quantities of incompletely separated fibers. These coarse fibers, also referred to as shives or fiber bundles, are retained on a 28 mesh screen during a Bauer/McNett classification. The coarse fibers are typically weak, have poor bonding characteristics~ and contribute to poor print quality in the final sheet of paper.

Existing processes for removing the coarse fraction from mechanical pulp utilize either perforated or slotted screen baskets. The rejected stock from these screens is returned to a reject refining system wherein the coarse fraction is rendered acceptable by further refining. ~;~
This method for ~eparating coarse fiber from the mechanical pulp for further treatment is inefficient.

Coarse fraction removal efficiency is a function of the mass reject rate and the screen configuration. As the mass reject rate increases, a larger portion of the coarse fraction in the feed stock is rejected. At mass reject rates of 25%, the removal efficiency of the coarse fibers approaches 50%. Even with extremely fine slotted screen baskets increases in the system removal efficiency to about 60% require an increase in the mass reject rate to : . ~

- , : :

. , ., .

2~8~

Docket No. 40010-1002 over 30%. Thus, high removal efficiencies are achieved at the expense of overloading the reject refining system.

Therefore, it has become apparent that a need exists for an improved method for fractionating mechanical pulp which results in a significantly higher coarse fraction removal efficiency at a significantly lower mass reject rate.

Summary of the Invention The present invention is an improved method for removing coarse fibers from mechanical pulp which utilizes a rotating disk filter, also known as a spray disk filter.
The use of rotating disk filters in the paper pulp processing industry has previously been known, however, their use has been designed to remove impurities such as -~
agglomerated inks from the pulp by carrying the impurities through the filter. See International Application WO
90/06396. The acceptable fraction of the pulp remains on the filter with the impurities being rejected by passing through the filter with a waste stream.

The method of the present invention works directly ~ ~;
opposite. Pulp stock at a controlled consistency is sprayed against the rotating disk filter through a plurality of spray nozzles positioned perpendicular to the disk. The accepted stock passes through the fabric and is collected in ;
a tank or transported for further use. The rejected stock ~ ' : . ~

Docket No. 40010-1002 remains on the disk and is thrown off the fabric by the centrifugal force generated by the rotating disk and is collected separately.

Removal efficiency of this method is a function of S mass reject rate, which is dependent on the fabric mesh size covering the disk, the feed stock flow and consistency, and the spray nozzle pressure.

Accordingly, it is an object of the present invention to provide a method for fractionating mechanical pulp resulting in significantly higher coarse fraction removal efficiencies at significantly lower mass reject rates.

These and other features and advantages of the present invention will be better understood by reference to the following detailed description, the accompanying drawings, and the appended claims.

Brief Description of The Drawings Fig. 1 is a perspective view of a spray disk filter useful in the present invention;

Fig. 2 is a perspective view of the spray disk filter of Fig. 1 including the drive mechanism and the pulp feeding system;

.. ... . . .. .. . . ~

2 3~

Docket No. 40010-1002 Fig. 3 is a perspective view of a plurality of spray disk filters incorporated as a processing system;

Fig. 4 is a flow diagram of one filtration system ;
in accordance with the present invention in which the accepts fraction is filtered twice; and : ': ': ~' Fig. 5 is a flow diagram of another filtration system in accordance with the present invention in which the rejects fraction is filtered twice and the accepts fractions are combined.
:':
Detailed Description -~

An improved method for fractionating mechanical pulp has been developed using a spray disk filter system, -generally designated 10, as shown in Fig. 1. Fig. 1 illustrates two spray disk filters A and B arranged in a back-to-back relationship, each operating in an identical manner. The spray disk filter 10 is a conventional apparatus manufactured by the Celleco Corporation.
Hereinbelow filter A will be described. It is to be understood that the description of filter A pertains similarly to all filters.
.:
Filter A consists of a disk 12 with its inside face covered by a fabric screen 14. Lignocellulosic pulp, and more particularly mechanical pulp stock 16, at a controlled consi~tency, is sprayed against the inner faces ~ Q ~

Docket No. 40010-1002 of disk 12, which is rotating, through a plurality of stationary spray nozzles 18 oriented perpendicular to the disks. A first fraction of the pulp passes through the filter and constitutes an accepts fraction and a second fraction of the pulp is retained on the filter and constitutes an enriched concentration of coarse fibers and constitutes a rejects fraction. The coarse fraction removed will typically range from +14 to +28 mesh.

As shown in Fig. 2, the fabric covered disks 12 are mounted on a center drive shaft 20. The disks are typically about eight feet in diameter with the outer 18 inches of the radius being covered by fabric 14. The fabric screens are usually made out of nylon or woven stainless steel. In one commercially available design, the fabric screen is divided into sectors individually mounted around the outer radius of the disk by support members 22. By utilizing sections of fabric, the screen can be easily installed and sections can be replaced when necessary.
Screens are commercially available from the filter manufacturer.

The center shaft 20 is coupled to a variable speed motor 24 (typically 25 horsepower) via a drive assembly 26 which rotates the center shaft therefore turning both disks at the same speed. Normal operating speeds depend upon a number of factors including pulp slurry consistency and the diameter of the disks, with the speeds ranging from about 10 ~ 2-.,33 Docket No. 40010-1002 to about 60 rpms. A preferre~ operating speed for an eight foot diameter disk is about 33 rpms.

The disks 12 and the shaft assembly are enclosed within a housing 28 with the motor 24 and drive assembly 26 mounted outside the housing. A removable lid (not shown covers the upper portion of the housing 28 and provides access to the filters for maintenance.

Mechanical pulp stock 16 (Fig. 1) is introduced into headers 30 through feed ports 32 located in the housing 28. A midfeather 31 separates the headers 30. The pulp 16 typically has a consistency ranging from about 0.30 to about 1.50 percent of pulp to water and preferably from about 0.5 to 1.0 percent.

The headers 30 feed a manifold of nozzles 18 -located on the surface of the headers. In one commercial design, on each header there are located between 44 and 58 full-cone nozzles each having a one-half inch diameter discharge orifice 37 (Fig. 1). The nozzles are oriented at to the disks, however the orientation can be varied such that the nozzles are positioned at an angle of about 40 to about 130 with respect to the direction of the disks 12.
Pulp stock is sprayed into chamber 11 against the rotating inner face 13 of fabric screen 14 through the nozzles 18 under pressure ranging from about 10 to about 35 psig and preferably 15 to 20 psig. The pulp is sprayed against the fabric screen 14 at a flow rate ranging from about 14 to ~t~ 7 Docket No. 40010-1002 about 48 gpm/ft2 and preferably 23 to 36 gpm/ft2. A first fraction of the pulp passes through the fabric screen 14 and constitutes an accepts fraction 34 and a second fraction of the stock is retained on the fabric screen and constitutes an enriched concentration of coarse fibers and constitutes a rejects fraction 36.

The accepts fraction 34 of the stock passes through the fabric to the outside of the disk 12 where it is collected in a chamber 38 located behind the disk 12.
Chamber 38 is separated from the chamber 11 by the fabric screen 14 and a seal (not shown). The rejects fraction 35 which does not pass through the fabric is retained and removed from the fabric screen 14 via a combination of centrifugal force and a cleaning shower described below.
lS The accepts fraction 34 contained in chamber 38 flows by gravity out of the chamber through an exit conduit 40. The rejects fraction 36 after being thrown off the screen flows by gravity to the bottom of the chamber 11 exits the housing through a rejects chute 42 for further processing or disposal through rejects conduit 44.

The spray disk filter is preferably also equipped with cleaning showers 46. There are two high pressure fabric cleaning bars 47 located on the feed side and the outlet side of each disk. These showers act to spray water and remove any fiber that may tend to clog the fabric. The showers on the inner feed side 13 are oriented at 90 to the fabric and will flush through the fabric to the outside of .::;: .; .

Docket No. 40010-1002 the disk. The showers on the outlet side 15 are oriented at 90 to the fabric and will spray towards the feed side. Two :
intermittent high pressure, rotating cleaning showers (not shown) are loca~ed in the top of the headers. These spray for ninety seconds on one hour intervals and act to keep the internal surfaces of the headers 30 clean.
::
The mesh size of the fabric, the feed stock consistency and flow, and the nozzle pressure are all controlled to prevent formation of a pulp mat on the fabric screen 14.
Feedstock consistency range~ and nozzle pressures ranges have been disclosed above. Several different media including nylon or woven stainless steel can be utilized having a mesh ranging from about 300 to about 1000 microns, with a preferred mesh ranging from about 450 to about 800 microns. ~igher consistency results in higher removal efficiency. Since the fabric screen 14 acts as a barrier, for a given disk area, disk speed and stock flow rate, higher consistencies result in more fiber being present at the fabric s~rface at any given point in time. At higher ~;
consistencies a thicker mat will form on the fa~ric. The mat will act as a barrier and will further filter out coarse fibers. While a higher stock consistency is desirable, it is limited because as the consistency increases, the mass reject rate also increases.

Other factors affecting coarse fraction removal efficiency are nozzle pressure' feed stock flow rate, and ~':

2 ~

Docket No. 40010-1002 feed stock coarse fraction content 3% to 10% of total fiber.
These factors appear to have less impact on separation efficiency than media size and feed stock consistency when in the ranges previously discussed.

Disk speed is another factor affecting coarse fraction removal efficiency. Disk speed impacts efficiency in one of two ways. At high feed stock consistencies, higher disk speeds aid in minimizing mat formation by increasing the centrifugal force used to remove the coarse fraction from the feed side of the disks. Disk speed can be used to fine tune the effect of media size through the concept of apparent aperture modification. At higher disk speeds, any opening appears smaller. This effect is expected to be minimal because of the relative velocities of the disk and the stock flow.

Fig. 2 illustrates the present invention using two disks, however the invention can incorporate any number of -disks. For example, a plurality of disk filtration systems 60, 62, 64, 66 each corresponding to the system shown in Fig. 2 may be mounted on a single drive shaft 20 as shown in Fig. 3. The mode of operation of the filters in Fig. 3 is identical to that discussed with reference to Figs. 1 and 2.
The accepts fraction 34 is collected from the system in a stock chest 46, with the rejects fraction 36 being collected in a separate stock chest 48.

,.~ ; . , :- , , : ~ , :. : .

Docket No. 40010-1002 For even higher coarse fraction removal efficiencies, the accepts fraction 34, and the rejects 36 fractions can be filtered a second time as schematically illustrated in Figs. 4 and S respectively.

S In Fig. 4 the mechanical pulp stock 16 is sprayed onto a first rotating disk filter or plurality of filters 50 such that a first fraction 34 passes through the filter~s) and constitutes an accepts fraction A and a second fraction 36 is retained on the filter(s) and constitutes a rejects fraction R. The accepts fraction A is conveyed to a second rotating disk filter(s) 52 such that a third fraction 54 of the pulp passes through the filter(s) and constitutes a second accepts fraction A' and a fourth fraction 56 of the pulp is retained on the second filter(s) 52 and constitutes a second rejects fraction R'. In implementing this system, ~
the operating conditions of the first and second filtration -systems can be the same but will preferably be different and optimized to provide the best balance of pulp quality and operation efficiency. Thus, the two filtration systems may employ different filter media, meshsize, consistency, rotational speed, etc. Furthermore, in a commercial setting first filtration systems may be a first bank of 1 to 8 filters and second filtration system 52 may be a second bank of 1 to 8 filters mounted on the same or a different drive shaft.

Alternatively, in Fig. 5 the mechanical pulp stock 16 is sprayed onto a first rotating disk filter(s) 50 such Docket No. 40010-1002 that a first fraction 34 passes through the filter(s) and constitutes an accepts fraction A and a second fraction 36 is retained on the filter(s) and constitutes a rejects fraction R. The rejects fraction _ is then conveyed to a ~.
second rotating disk filter(s) 52 wherein it is sprayed on the filter(s) such that a third fraction 54 of the pulp passes through the second filter(s) and constitutes a second ~ -accepts fraction A' and a fourth fraction 56 is retained on the second filter(s) and constitutes a second rejects ~:
fraction R'. Normally, the second station of filters can :~
contain a smaller number of filters because once the stock has been filtered a first time, a lesser amount of stock remains to be filtered for the second time.

The spray disk filter system of the present invention has resulted in significantly higher coarse fraction removal efficiencies at significantly lower mass rejects rate. Removal efficiencies of 80 to 95% are achievable at system mass reject rates of 15 to 60%.
Removal efficiency is defined as: .
(shives fed - shives accepted) X 100 (shives fed) and mass reject rate defined as:

~fiber fed - fiber acceDted) X 100%
~(fiber fed) 2 ~

Docket No. 40010-1002 The accepts from the filter are more dilute than the feed (have a lower consistency) for both of the processes depicted in figures 4 and 5. No further consistency adjustment is necessary for secondary processing of the accepts such as in figure 4. The rejects, however, are more concentrated (have a higher consistency) for both processes depicted in figures 4 and 5. Depending on the secondary media size, it may be necessary to further dilute the rejects stream before processing as in figure S. A
recommendation is to dilute to a consistency similar to that of the first disk feed consistency if both disks are equipped with the same size media.

The invention has been disclosed with respect to its preferred parameters in which test runs have been performed with the following non-limiting examples~
~:

Example #l A prototype spray disk filter with a single 3 foot diameter disk and 3 inlet nozzles was used. In theory, the -degree of fractionation can be affected by several operating variables including mesh size, nozzle angle, disk speed, nozzle pressure, and pulp consistency. Two series of trials were performed to evaluate the capability of the spray disk filter to fractionate mechanical pulp. The variables 2 ~ ? ~

Docket No. 40010-1002 evaluated and the ranges evaluated are summariæed in Table Table 1 Spray Disk Filter Trial Conditions ~;
5June 26-27 Au~ust 14-15 ~ -Media Size ~micron)600, 800 6 1000300, 480, 540 600, 800 and 10~0 Consistency (%) 0.60 - 1.54 0.28 - 1.14 Number of Nozzles 1 3 Nozzle Angle (degree) 90 9~
Nozzle Pressure (psig) 25 15 - 30 Disk Speed (rpm) 96 96 Mechanical pulp stock from a refined pulp surge chest was shipped for trials. A total of 21 trial runs were conducted during two trial periods (June 26-27 and August 14-15, 1991). Trial runs were conducted such that three different system configurations were evaluated. In the first, the spray disk filter was used as a single stage screening device. Two 2-stage configurations were also evaluated. Stock was screened through a primary stage spray disk filter and the accepted stock was screened through a second stage spray disk filters as shown in Fig. 4. In the second approach, the rejected stock (coarse fraction) from the primary spray disk filter was screened through a secondary spray disk filter as shown in Fig. 5.

~ 7 Docket No. 40010~1002 Flow rates and consistencies were measured for each trial run. Samples were collected from the feed, accept and reject for Bauer/McNett fiber classifications.
During the trials, freeness was measured on the feed, accept and reject streams. The trial conditions, fiber classifications, mass balance, mass reject rate, volume reject rate and efficiency calculations for each trial run are summarized in Table 1.

In general, +28 fraction removal efficiency, mass reject rate, and volume reject rate all decrease with increasing media size. Consistency has the opposite effect.
Higher consistencies result in higher efficiency, higher mass reject rate and higher volume reject rate. When flow rate or +28 fraction are included in the models for mass reject rate, volume reject rate or efficiency, they have less effect on the performance of the filter.

Docket No. 40010-1002 Table 1 SYstem 1 (Fig. 4) Primary Sta~eSecondarY Stage Media 480 micron480 micron Feed Consistency 0.25%
Feed +28 Fraction 13.3%
Nozzle Pressure 15 psig 15 psig Flow Rate 100 gpm System Mass Reject Rate 34.3~
System +28 Removal Efficiency 97.9%
Primary Stage Secondary Sta~e Media 800 micron540 micron Feed Consistency 0.50%
Feed +28 Fraction 13.3%
Nozzle Pressure 20 psig 20 ~sig Flow Rate 100 gpm System Mass Reject Rate 25.6%
System +28 Removal Efficiency 91.6%

Svstem 2 (Fig. 5) Primary Stage SecondarY
Stage Media 480 micron480 micron Feed Consistency 0.75%
Feed +28 Fraction 13.3%
Nozzle Pressure 15 psig 15 psig Flow Rate 100 gpm System Mass Reject Rate 13.7~
System +28 Removal Efficiency 80.7%

: :

2 ~ ? 2 ~

Docket No. 40010-1002 Example #2 A commercial size spray disk filter unit with two 7'5" diameter disks and 112 spray nozzles was evaluated in a mechanical pulp mill. Operating variables studied were mesh size, disk speed, nozzle pressure, pulp consistency, and feed flow rate. Three series of trials were conducted February 13, 1992 to July 31, 1992. The variables studied are summarized in the table below.

Pilot Plant SpraY Disk Filter Trials Conditions Media Size (microns~ 450, 600, and 800 Feed Consistency (%) 0.43 - 1.46 Nozzle Pressure (psi) 10 - 30 Disk Speed (rpm) 25 - 41 Feed Flow Rate (gpm) 471 - 1454 The pilot plant unit was installed such that it would be operated in three different configurations: 1) both disks fed in parallel; (2) series operation with disk 1 accepts (fine fraction) feeding disk 2; (3) series operation with disk 1 rejects (coarse fraction) feeding disk 2.

The installation included feed flow measurement and control for both disks, feed consistency measurement and control for both disks, and in-line pressure measurement for ~ -the feed to both disks. The level in the intermediate storage tank was measured and controlled. Combined accepts flow rate was measured in-line.

Docket No. 40010-1002 Pilot plant trials were conducted in which the system operated in configuration 3: primary disk coarse fraction was filtered a second time by the secondary disk.
Flow rates and consistencies were measured during each trial. Samples were collected from disk 1 feed, disk 2 feed, disk 1 accepts, disk 2 accepts, disk 1 rejects, disk 2 rejects, and combined accepts. These samples were then classified on a Bauer/McNett classifier and the freeness was measured using the Canadian Standard Freeness test. Results are summarized in the following table.

Primary Disk Secondarv Disk Media (microns) 450 450 Feed Consistency (%)0.46 - 0.620.44 - 0.74 Feed Flow Rate (gpm)914 - 1454 593 - 809 Feed +28 Fraction (%)3.38 - 7.175.66 - 12.4 Nozzle Pressure (psi)10 - 30 10 - 30 Speed (rpm) 25 - 41 25 - 41 Disk Mass Reject Rate (%) 47.9 - 75.0 58.6 - 82.0 Disk Removal Efficiency (%) 81.5 - 97.3 95.8 - 98.6 System Mass Reject Rate (%) 35.7 - 59.3 ~ d~
System Removal Efficiency (%) 81.7 - 93.5 : . ,:
:: ': ' ~ -Primary Disk Secondary Disk Media (microns) 600 600 Feed Consistency (%) 0.43 - 0.73 0.49 - 0.77 Feed Flow Rate (gpm) 957 - 1435 482 - 1053 Feed +28 Fraction (%) 3.42 - 9.29 6.32 - 9.54 Nozzle Pressure (psi) 10 - 30 10 - 30 Speed (rpm) 33 Disk Mass Reject Rate (%) 35.8 - 78.7 54.1 - 86.2 -~
17 ~-~" . . .:

: :: ~ :::

: . ' '. . . ' .: . . , ~ ~Q~ ~1 Docket No. 40010-1002 Disk Removal Efficiency (~) 76.3 - 98.8 86.3 - 98.6 System Mass Reject ~ate (~) 2902 - 65.5 System Removal Efficiency (~) 72.0 - 97.1 Primary Disk SecondarY Disk s Media (microns) 800 800 Feed Consistency (%) 0.85 - 1.28 0.65 - 1.46 Feed Flow Rate (gpm) 695 - 1236 471 ~ 934 Feed +28 Fraction (%)3.66 - 7.09 4.74 - 8.48 Nozzle Pressure (psi) 10 - 30 10 - 30 Speed (rpm) 33 - 41 Disk Mass Reject Rate (%)62.2 - 78.451.1 - 90.3 Disk Removal Efficiency (~) 85.9 - 93.8 83.5 - 98.3 System Mass Reject Rate (%) 28.2 - 6~.2 System Removal Efficiency (%) 56.1 ~ 89.1 In general, +28 fraction removal efficiency, mass reject rate, and volumetric reject rate all decreased with increasing media size. Higher consistencies, however, had the opposite effect; +28 removal efficiency, mass reject rate, and volumetric reject rate all increased with increasing consistency. Flow rate and disk speed were found to have a less significant effect on efficiency, mass reject rate, and volumetric reject rate.

It should be noted that the fiber source for this mechanical pulp mill was aspen. Optimum filter operating conditions are likely to vary for different fiber sources depending on individual fiber characteristics.

, ~... , .

. .
. ~ , ,. : . .. .
.

~ 1 n ,~

Docket No. 40010-1002 The preceding description has been presented with reference to a presently preferred embodiment to the invention shown in the drawings. Workers skilled in the art and technology to which this invention pertains will appreciate that alterations and changes in the described structure can be practiced without departing from the spirit, principles and scope of this invention.

Claims (20)

1. A method for removing coarse fibers from a lignocellulosic pulp comprising the steps of:
spraying a lignocellulosic pulp slurry containing said coarse fibers against a rotating disk filter such that a first fraction of said pulp passes through said filter and constitutes an accepts fraction and a second fraction of said pulp is retained on said filter and contains an enriched concentration of said coarse fibers and constitutes a rejects fraction; and removing said accepts fraction and said rejects fraction from said filter.

2. The method of claim 1 wherein said pulp is a mechanical pulp.

3. The method of claim 2 wherein said disk filter is positioned with its center axis horizontally oriented and said pulp is sprayed on said filter in a horizontal direction.

4. The method of claim 3 wherein said disk filter is formed from a filter media having a pore size ranging from about 300 to about 1000 microns.

5. The method of claim 4 wherein said pulp slurry has a consistency ranging from about 0.1 to about 2.0 percent.

6. The method of claim 5 wherein said disk filter rotates at about 10 to about 60 rpm.

7. The method of claim 6 wherein said pulp slurry is sprayed against the disk filter under pressure ranging from about 15 to about 35 psig.

8. The method of claim 7 wherein said pulp slurry is sprayed against the disk filter through nozzles oriented at an angle of about 40 to 130- with respect to the direction of rotation of said filter.

9. The method of claim 8 wherein the said pulp slurry is sprayed against the disk filter at a flow rate ranging from about 14 to about 48 gpm/ft2.

10. The method of claim 9 including the additional steps of:
conveying said accepts fraction to a second rotating disk filter;
spraying said accepts fraction against said second rotating disk filter such that a third fraction of said pulp passes through said filter and constitutes a second accepts fraction and a fourth fraction of said pulp is retained on said second filter and contains an enriched concentration of said coarse fibers and constitutes a second rejects fraction.

11. The method of claim 9 including the additional steps of:
conveying said rejects fraction to a second rotating disk filter;
spraying said rejects fraction against said second rotating disk filter such that a third fraction of said pulp passes through said filter and constitutes a second accepts fraction and a fourth fraction of said pulp is retained on said second filter and contains an enriched concentration of said course fibers and constitutes a second reject fraction.

12. The method of claim 3 wherein the method is conducted under conditions which provide a removal efficiency ranging from about 80 to about 95 percent at a total system mass reject rate ranging from about 15 to about 60 percent.

13. The method of claim 12 wherein the coarse fibers removed are greater than 28 mesh.

14. A method for removing coarse fiber from mechanical pulp comprising the steps of:
spraying a mechanical pulp slurry against a first rotating disk filter;
passing a first accepts fraction of said pulp slurry through said first disk filter;
retaining a first rejects fraction of said pulp on said first disk filter;

conveying said first accepts fraction to a second rotating disk filter;
spraying said first accepts fraction against said second rotating disk filter;
passing a second accepts fraction of said pulp through said second disk filter;
retaining a second rejects fraction on said second disk filter;
collecting said second accepts fraction; and removing said first rejects fraction and said second reject fraction from said disk filters.

15. A method for removing coarse fiber from mechanical pulp fraction comprising the steps of:
spraying a mechanical pulp slurry against a first rotating disk filter;
passing a first accepts fraction of said pulp slurry through said disk filter;
retaining a first rejects fraction of said pulp on said first disk filter;
removing said first rejects reaction from said first disk filter by centrifugal force through a reject chute;
conveying said first rejects fraction to a second rotating disk filter;
spraying said first rejects fraction against a second rotating disk filter;
passing a second accept fraction through said second disk filter;

retaining said second rejects fraction on said second disk filter;
removing said second rejects fraction from said disk filter through a reject chute.

16. The method of claim 1 wherein said method is performed using a plurality of disk filters mounted for rotation on a common drive shaft.

17. The method of claim 10 wherein said method is performed using a plurality of first and second disk filters mounted for rotation on separate drive shafts.

18. The method of claim 11 wherein said method is performed using a plurality of disk filters mounted for rotation on separate drive shafts.

19. The method of claim 10 wherein said second accepts fraction is filtered a third and fourth time.

20. The method of claims 11 wherein said second rejects fraction is filtered a third and fourth time.