PEER-REVIEWED PAPERMAKING

Determination of drainage resistance

coefficients under known shear rate
MARK A. PARADIS, JOSEPH M GENCO, DOUGLAS W. BOUSFIELD, JOHN C. HASSLER, AND VAUGHN WILDFONG

ABSTRACT: The drainage rate of papermaking stock determines the design and operation of the wet end
of a paper machine. For single and twin wire machines with various elements, the effectiveness of the drain-
ing process cannot be quantified unless there are good numbers for the drainage resistance of a given pulp
under conditions that resemble the industrial process. A recently developed laboratory drainage device can
estimate the viscous drainage resistance at short times and high drainage rates. However, this device has no
shear on the pulp during drainage. This paper presents a new method to measure the drainage rate under
known shear conditions. We report the influence of shear rate and various constituents in lightweight coat-
ed (LWC) paper on drainage rate. We also investigated the effect of shear rate on drainage resistance.
Application: Design of paper machine formers, drainage rates on existing paper machines, headbox
slurry characterization.

ulp drainage rates and viscous resist- on mat thickness into the models.
P ance coefficients are a function of the
physical properties of the paper furnish.
short time drainage results, but the time
scale and conditions were long compared
to that of an actual paper machine. The
The largest short-coming of the
method described by Wildfong et al. [8]
Important factors influencing wet end reported drainage resistance was not con- is that the pulp suspension is not
drainage are pulp composition, average stant; it increased as the mat thickness exposed to shear during drainage. Hung
fiber length, fiber length distribution, [Link] result pointed to the need et al. [10] describe a donut-shaped piston
fines content, charge level on the stock, for a short time and thin mat test. cylinder apparatus under development
and degree of stock hydration. Stock Wildfong et al. [8, 9] described a for determining the permeability of pulp
drainage also depends on the chemical rapid drainage tester and showed that the fiber under simulated forming condi-
processing aids such as starch, coagu- results can help predict pilot scale data. tions. A transducer in the piston will
lants, flocculants, and wet-strength addi- The test used a laser triangulation sensor serve to measure the flow rate through
tives and the level of shear present on the to measure the motion of the liquid free piston displacement and pressure. Shear
wet end of the paper machine. Drainage surface during drainage. The test was is applied by rotating the wire on the
rates have been difficult to quantify at rapid with drainage rates that matched inside surface of the device. This device
industrial significant conditions. industrial [Link] resistance coef- has the potential to characterize short
Standard methods to characterize ficient again increased as the mat formed, time rapid dewatering under shear, but
drainage rates, such as the Canadian although in filtration theory, this coeffi- we have not seen reports of results so far.
Standard Freeness (CSF), are good for cient should be constant. The drainage Arslan et al. [11] investigated a
quality control and comparison with resistance should increase with the mat method to measure retention at a known
other stock, but these tests do not pro- thickness, but the specific cake resistance shear [Link] used an inverted cone to
vide usable fundamental parameters for should be essentially constant except generate a known and uniform shear rate
models that predict pressure distribu- when there is a very large increase in in the region over the wire in an appara-
tions and drainage rates during forming. pressure drop across the mat, which tus that resembles a cone-and-plate
Models for pressure distribution, as would not be expected on the wet end of rheometer. The drainage region is meas-
described by Zhao and Kerekes [1], a paper machine. ured at a known and uniform shear rate
Zahrai and Bark [2], Roshanzamir et al. Fines retention plays an important during a retention test. They compare
[3], Green [4], Green et al. [5], and role in the change in resistance coeffi- results to the standard Britt jar and report
Dalpke et al. [6], all depend on some cient with time. Certain grades, such as them for different systems. Arslan et al.
parameter—often called drainage resist- newsprint, show a rapid increase in the also report some drainage rate data [11]
ance coefficient or a Darcy law coeffi- coefficient while others show a moderate measured over long time periods that do
cient—that relates the dewatering rate increase. Other tests suggest that the not match industrial drainage rates.
with the pressure. In theory, this parame- increase in the resistance coefficient is a In this work, we describe a new
ter should be a standard measurement of result of fine material being trapped in drainage apparatus that permits accurate
pressure drop across a mat of pulp. In different layers of the forming mat, caus- estimation of the drainage resistance of
fact, this coefficient depends on basis ing that mat to have different viscous headbox and other papermaking stock
weight and other local conditions and is resistance values in the different layers. In samples under industrially significant
difficult to measure for conditions that spite of the changing resistance coeffi- conditions. The new apparatus builds
mimic the high speed and short-time cient, we should be able to use the aver- upon the previous work of Wildfong et
drainage conditions of a modern paper age values in the models above to obtain al. [8] and Arslan et al. [11].The drainage
machine. Mantar et al. [7] reported some good predictions or build a dependence rate is determined by a conductance

12 TAPPI JOURNAL AUGUST 2002

 PAPERMAKING

Variable Electronic C
measurement in the stock remaining in the shear [Link] report speed motor unit
10 mv
results for a number of pulp suspensions and for different shear
Electrodes
rates. D

EXPERIMENTAL Upper
The theory and basic experimental apparatus are described by chamber Computer

Support stand
Wildfong and coworkers [8], while the shear cell is described by
Arslan [11]. Darcy’s law can be modified to define the drainage
resistance (a) of a paper stock as it is being filtered on the wet
end of a paper machine. Forming Pressure
fabric transducer

(1) Forming Quick Reservoir

fabric opening
support valve
In the above equation (∆Pm) is the pressure drop across the
fibrous mat that is [Link] have defined the thickness of the Vacuum
mat to be (L) when the superficial drainage velocity is (v). To Bleed pump
valve
estimate the drainage resistance (a), we must construct an appa-
ratus that simultaneously measures the pressure drop (∆Pm), the Filtrate
mat thickness (L), and drainage velocity (v) in real time. recepticle
Drainage apparatus
The drainage apparatus (F Fig. 1) consists of an upper chamber
equipped with a forming fabric, fabric support, and pressure
transducer located under the forming [Link] pressure trans-
ducer gauges the differential pressure across the mat and the
forming fabric.A pair of electrodes estimates the volume of the
slurry remaining in the shear cell in real time. The differential
pressure across the mat and forming fabric is induced by a vac-
uum pump, which is connected to the filtrate receptacle using a 1. Schematic of new drainage apparatus.
large reservoir that functions as a capacitance in the system. A
quick opening value separates the upper chamber and the fil-
trate receptacle. In the experiment a known vacuum is drawn (3)
on the stock, which is allowed to drain very rapidly through the
forming fabric. The total pressure drop (∆Pt) is measured as a function of
The upper chamber uses a cone-and-plate rheometer to time using a pressure transducer shown in Fig. [Link] pressure
apply a quantifiable shear field to fluid as it passes through the drop across the forming fabric (∆Pw) is estimated as a function
thickened mat and the filtration media. Therefore, the drainage of drainage velocity from data obtained from a separate experi-
takes place under a known and uniform shear rate. The signals ment in a flow loop.
from the electrodes and the pressure transducer are recorded by The mat thickness (L) is estimated from the drainage vol-
a computer in real [Link] data collection frequency is 500 Hz. ume (V), the retention coefficient (R), the drainage area (A) and
We use the data to estimate the drainage velocity (v). Drained liq- the properties of the stock:
uid is analyzed to determine the overall retention coefficient
(R), while the material remaining in the drainage apparatus can
be tested for water retention value (WRV), composition, and
fiber length. Depending upon the vacuum in the apparatus, the (4)
entire drainage experiment takes only a few seconds so that the
drainage velocities are similar to those present on high speed where Co and ρs are the original consistency and density of the
commercial paper machines. sample entered into the test apparatus. To estimate the mat
Since drainage experiments are conducted under constant thickness, values are required for the void fraction (εm) and den-
area (A) the drainage velocity (v) is related to the change in the sity (ρp) of the particles comprising the mat or wet web. Lastly,
volume (V) in the apparatus as a function of time (t). the drainage resistance (a) can be estimated from the appropri-
ate measurements as a function of time by rearranging equation
(1) and by appropriate substitution.
(2)

The pressure drop across the mat (∆Pm) is related to the

total pressure drop (∆Pt) and the pressure drop across the form-
ing fabric (∆Pw).
(5)

VOL. 1: NO. 6 TAPPI JOURNAL 13

PAPERMAKING

Headbox sample # grade (1) (2) (3)

Bond Highly filled LWC
opaque book grade basesheets
CANADIAN
Furnish components 85% HW 50% Sulfide 37% SWK STANDARD
15% SW 20% SWK 24%TMP COMPONENT FREENESS (mL)
PCC 30% HWK 36% GW Stone groundwood 39
40% added broke 3% BCTMP TMP 64
160ppt PCC 15% added ctd. broke Coated broke 112
100ppt TiO2 BCTMP 120
Refined bleached SW kraft 649
Headbox consistency, % 0.43 0.39 0.47 Headbox 40
Headbox fines content, % 32.3 36.0 44.0
Headbox ash content, % 20.6 24.4 8.30 II. Freeness values for components in
Freeness, csf 355 103 40
lightweight coated (LWC) paper.
I. Characteristics of headbox samples.

As the mat thickness increases with time, the resistance stock used to produce lightweight coated (LWC) publication
increases to reduce the drainage rate. For materials such as solid [Link] chose these grades because they have a wide range of
particles of uniform size, the drainage resistance is constant and pulps and fines content (T Table I). The headbox sample for the
is not a function of time. However for pulp suspensions, the LWC paper and the highly filled opaque grade came from two
resistance coefficient often increases with basis [Link] vol- local paper [Link] sample for bond paper was taken from the
ume drained converts to basis weight (BW) as: headbox of the experimental fourdrinier paper machine at the
University of Maine during papermaking experiments. Table I
summarizes the characteristics of the three headbox samples
slurries. Filler content determined using a 550oC combustion
(6) test ranged from a low of 8.3% for the LWC to 24.4% for the
highly filled opaque book grade. Similarly, the total fines content
We estimated the pressure drop across the forming fabrics ranged from 32% for the bond paper to 44% for the LWC paper.
in an external flow loop using water shown in Fig. [Link] flow We also used the apparatus to determine the drainage
loop consists of a reservoir for holding 20 gal of water, a sub- resistance for the various pulp components that go to the blend
mersible pump, 1 in. plastic pipe, a rotometer for estimating the chest in the production of LWC publication paper (T Table II). In
flow rate, fabric holder, two manometers open to the atmos- this example, the constituents in the LWC basestock were stone
phere, and associated valves and fittings. In operation, a sample groundwood, thermomechanical pulp (TMP), refined softwood
of the forming fabric is contained within a fabric holder and kraft pulp, bleached chemi-thermomechanical pulp (BCTMP),
water is pumped from the 20 gal tank through the flow meter and coated [Link] II gives the freeness values for the com-
and then through the forming [Link] pressure drop across ponent pulps used in the production of the LWC publication
the fabric (∆Pw) is measured with the two manometers located paper. CSF values ranged from 39 mL for the stone groundwood
upstream and down stream of the fabric while the superficial pulp to 649 mL for the refined bleached softwood kraft pulp.
fluid velocity is estimated from the reading taken using the Determining the drainage resistance of the component pulps
flowmeter and the dimensions of the 1-in. pipe. allowed us to compare the drainage resistance of the headbox
The data are plotted as pressure drop across the forming sample which had a CSF value of about 40 mL.
fabric (∆Pw) versus velocity (v). These data are subsequently The cone used in the drainage experiments had an angle of
used to correct the total pressure drop (∆Pt) according to Eq. (3) 10o and ran at speeds of 0, 250, 500, 750, and 1000 [Link] cor-
for each value of the estimated drainage velocity measured in responded to shear rates of about 0-594 s-1 and shear stresses of
the drainage apparatus. 0-0.594 Pa assuming the stock had a viscosity of water. The
For cone and plate type rheometers with small cone angles, experiments to characterize the headbox and LWC pulps were
the shear rate applied to the fluid can be calculated by the fol- only conducted at 500 [Link] drainage area in the device was
lowing equation: approximately 0.0127 m2 and corresponding to an inside diam-
eter of 12.7 cm. During the experiments, the vacuum setting in
the reservoir was set at 15 kPa.
(7) The flow resistance was estimated in the flow loop for a
variety of forming [Link] selected a triple layer publication
where w is the rotational speed and α is the cone angle. grade forming fabric and used it in the drainage experiments to
Experimental conditions estimate the flow resistance for the headbox and pulp samples.
The drainage resistance (α) was determined for a variety of sam- The publication grade forming fabric was 180x165 mesh in the
[Link] expect furnishes that have high values of the drainage machine- and cross-machine directions respectively. The form-
coefficient (α) and high water retention value (WRV) would ing fabric had a void fraction (εw) of 48.6% and an equivalent
also be expected to be difficult to dewater in the press and wire diameter (Dp) of 0.255 mm.
dryer section of the paper machine. The particle density (ρp) and the void fraction in the mat
We tested headbox samples for three different paper (εm) were estimated as described in Paradis et al [12]. the parti-
grades: bond paper, highly-filled opaque book paper, and base cle densities ranged from 1200 to 1450 kg/m3 depending on the

14 TAPPI JOURNAL AUGUST 2002

 PAPERMAKING

y(#1)=15x2+227.78x #4 (Double layer,

25,000 R2=0.9986 publication grades)
filler content. Values for particle density
and viod fraction in the mat are required y(#2)=8.6426x2+169.12x
R2=0.9849 #5 (Triple layer,
in Eqs. (3) and (5) to estimate the mat publication grades)
y(#3)=8.3416x2+71.782x
thickness and the drainage resistance. 20,000
R2=0.9984
y(#4)=30.628x2+354.49x
RESULTS R2=0.9987 #7 (Double layer,

DYNES/cm2
publication grades)
Forming fabrics flow resistance 15,000 y(#5)=32.571x2+178.25x
#1 (Single layer)
Figure 2 shows the results of the pressure R2=0.9938
#6 (Double layer,
drop measurements for a variety of form- y(#6)=15.088x2+186.46x publication grades)
R2=0.9968
ing fabrics. All of the pressure drop rela- #8 (Triple layer,
10,000 y(#7)=20.366x2+216.46x publication grades)
tionships shown in the figure follow a
R2=0.9988 #2 (Single layer)
quadratic equation, indicating that the
y(#8)=11.786x2+216.75x
inertial terms become important at high R2=0.9976 #3 (Single layer)
drainage rates. Forming fabrics with low 5000
values for the void fraction (εw) exhibited
the greatest flow resistance. These rela-
tionships were used later to estimate the
0
pressure drop of the forming fabric at a 0 5 10 15 20 25 30
given superficial drainage rate. cm/s
For flow in porous media over a
range of flow rates, the Ergun equation 2. Pressure drop across forming fabrics.
worked well (Bird et al. [13]
any mat that has [Link] pressure beneath the forming fab-
ric reaches the steady state value within about 0.2 [Link] volume
remaining in the shear cell decreases at a rapid [Link] rate of
drainage decreases as a mat forms on the wire. The important
experimental data are collected within about 2 s. Figure 5 illus-
(8) trates how the pressure changes beneath the wire as the
drainage velocity changes. The velocity is calculated by taking
In the Ergun relationship, ∆Pw is the pressure drop, ρ is the the difference in volume in the cell and dividing by the change
fluid density, Dp is the mean diameter of the wires comprising in time and the drainage [Link] drainage velocity at zero time
the fabric, εw is the void fraction of the fabric, Lw is the thickness is estimated from the pressure drop at that time and the wire
of the wire, Go is the mass flow velocity, and µ is the fluid vis- resistance.
cosity. We calculated the mean diameter (Dp) from the surface We ran six replicate drainage runs for each experimental
area per unit volume (Av) for the forming fabrics using the fol- condition and averaged the separate curves to obtain an average
lowing equation. [Link] increase in drainage velocity at short times is a result
of the time required to accelerate the fluid mass. As the mat
builds up on the wire, the drainage velocity [Link] coef-
(9) ficient of variation, or the percent difference between the indi-
vidual runs and the mean, is 15% to 20%. We estimated the
We approximated the parameter (av) from the surface area potential initial velocity, with no mat, from the data in Fig. 2 by
and volume of the forming wires comprising the [Link] did knowing the pressure drop [Link] start-up transient does
this by estimating the dimensions of the forming wires and by not invalidate the latter time data.
counting the strands under a microscope. Shear rate and the drainage resistance coefficient
When the experimental data were plotted in the form Figure 6 shows the effect of shear rate on the drainage resist-
of the Ergun equation, a line is obtained with a slope of 167 and ance coefficient (a) and ash retention. The figure plots the
an intercept of 1.78, which are close to the coefficients of 150 drainage resistance as a function of stirrer rpm for different basis
and 1.75 in Eq. (8). Figure 3 shows the results. The correlation weights of the mat that was formed (dry weight basis). We cal-
coefficient (R2) for the Ergun equation was 0.871, which indi- culated the shear rates from Eq. (7). They linearly depend on
cates a reasonable fit of the data. Considering the accuracy of rpm in which 0 to 1000 rpm relates to 0 and 594 [Link] stirring
measuring the above parameters, this result is excellent. This speed corresponded to a shear stress (t) of 0 to 0.6 Pa, assuming
result is a good validation of the Ergun equation and is a pow- that the slurry behaved as a Newtonian fluid and had a viscosi-
erful method to predict the resistance of a given fabric from the ty equal to that of [Link] estimate that 85% of the wire area
basic properties of the wire diameter, thickness, and void frac- sees this shear rate. As expected, the retention decreases with
tion. increasing shear rate.
Drainage cell Figure 6 shows that there is an increase in drainage resist-
Figures 4 shows typical output results from the drainage appa- ance with an increase in shear rate up to a maximum of about
[Link] about 1 s on the time axis, the valve is opened, and the 700 to 800 rpm. The maximum depended upon the weight of
pressure transducer sees a pressure drop across the wire and the fiber mat [Link] low mat weights (and thickness) there
VOL. 1: NO. 6 TAPPI JOURNAL 15
PAPERMAKING

100 0

PRESSURE, Pa
-5000

-10000

-15000

10 -20000

-25000
0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5

TIME, s
200

VOLUME, mL
150
1
0.001 0.01 0.1 1 100

50

0
0.5 0.7 0.9 1.1 1.3 1.5 1.7 1.9 2.1 2.3 2.5
3. Correlation of pressure drop across commercial forming
TIME, s
fabric.
4. Sample output from drainage device.

20000
14 LWC Basesheet
Theoretical Drainage Velocity Headbox Sample 18000 8 80
-1 * (DIFFERENTIAL PRESSURE)

12
∆P(t)
16000 Ash Retention
DRAINAGE VELOCITY, cm/s

7 70
14000
10
6 60
DRAINAGE VELOCITY, cm/s

12000
a for 50 gsm

ASH RETENTION, %
8
10000 5 50

6 8000
4 40
V(t)
6000
4
4000 3 a for 40 gsm 30

2
2000
2 20
0 0 a for 30 gsm
0 50 100 150 200 250 300 350 400
1 10
TIME, ms a for 20 gsm
0 0
5. Pressure and velocity at short times. Drainage rate at 0 100 200 300 400 500 600 700 800 900 1000

zero time is estimated from wire resistance and the STIRRER SPEED

applied pressure. 6. Effects of shear rate on drainage resistance and retention.

was little increase in the [Link] the basis weight increas- We used a value of 500 rpm for the remainder of the exper-
es, the increase in resistance is more pronounced. Even at high iments presented below. A rotational speed of 500 rpm gives
mat thicknesses, the resistance coefficient drops sharply above rise to a shear rate of about 298 s-1 and corresponds to a shear
a stirrer speed of 800 [Link] initial increase in drainage resist- stress of 0.3 Pa. Franzen [14] presents a numerical design model
ance with increasing shear rate may be a result of the increas- for predicting drainage on twin wire and top wire units.
ing force from the stirrer compressing the [Link] increase in Franzen’s model permits estimation of the shear rate and
resistance may be a result of the shear field that helps the fibers stresses present as the wet web passes over forming blades in
to deposit into a tighter [Link] decrease in drainage resist- top wire units. Numerical calculations performed for a
ance at the higher shear rates may be due to the disruption or newsprint furnish using the Franzen model suggest that using a
removal of the mat from the forming fabric. In some cases, this rotational speed of 500 rpm approximates the shear stress on a
disruption was evident at the end of the [Link] shear commercial top wire unit. Additionally, this stirring speed
rate or stress to disrupt the mat may be an important charac- proved to be convenient to use experimentally.
teristic of the wire and the pulp that can now be characterized Headbox samples
by these [Link] mat disruption may be a function of The drainage velocities for the three headbox samples investi-
the fiber type, retention system, filler content, and wire type. gated are summarized in Fig. 7 at a stirring speed of 500 rpm.

16 TAPPI JOURNAL AUGUST 2002

 PAPERMAKING

7 25
High filled
DRAINAGE VELOCITY, cm/s

6 Bond grade headbox sample book grade LWC

LWC basesheet headbox sample 20
5 High filled book grade headbox sample

a (g/cm3 s)x106
15
4

3 10

2
5
Bond grade
1

0
0 10 15 20 25 30 35 40 45 50 55
0 50 100 150 200 250 300 350 400
B/W, g/m 2
TIME, ms
8. Average drainage resistance coefficients for grades listed
7. Drainage velocity vs. time curves for headbox samples. in Table I.

25
Capillaries available for flow
Fine particle entrapment Headbox sample = 40
increasing drainage resistance 20 GW = 39 csf
TMP = 64 csf GW
Broke = 112 csf Headbox
a (g/cm3 s)x106

sample
15 BCTMP = 120 csf
Fiber mat SW = 649 csf

Forming
10 TMP
fabric
Broke

BCTMP
9. Schematic of the effect of fine particle plugging. 5
SW

There are small differences in consistency between the three 0

0 10 20 30 40 50 60 70
samples, but they vary only between 0.4% to 0.5%. All three of B/W, g/m
the headbox samples show the same increase in the velocity at
short times because the same forces are required to accelerate 10. Drainage resistance coefficients for furnish components
the fluid. However, at around 60 ms, significant differences are
for LWC base sheet.
apparent. The bond grade headbox sample shows a higher
drainage rate for the entire time when compared to the samples
for the highly filled opaque book grade or the LWC basestock. (εm), the surface area per unit volume (Sv) of the particles
This indicates that the mat formed is more open. Even at 200 trapped in the mat, and the viscosity (µ) as in the Carman-
ms, the bond grade sample has a higher drainage velocity, even Kozeny relationship as
though the mat formed at this point should be thicker than the
mat with the LWC sample or the book grade pulp.
Figure 8 illustrates the results after we converted the veloc- (10)
ity data into a resistance coefficient. The headbox sample for
the bond grade had a freeness of 355 mL compared to the 40 Figure 9 illustrates this [Link] bond grade sam-
mL for the LWC basesheet and the 103 mL for the sample for ple shows only a slight increase, but the other samples indicate
the highly filled opaque grade. Figure 8 shows that the bond an increase by a factor of [Link] bond grade sample has a low
grade stock had a much lower, and mostly constant drainage level of fine material.
resistance coefficient compared to the LWC and book grade. Figure 10 shows the results for the constituent pulp sam-
If the mat formed on the wire had constant physical prop- ples comprising the furnish going to the blend chest for the
erties, the drainage resistance would be [Link] increase LWC [Link] II gives the constituents and the freeness val-
in the drainage coefficients for all three samples shows that the ues for the pulps. The headbox sample shows the largest
characteristics of the mats change as drainage occurs. This is increase in the resistance coefficient while the softwood sam-
thought to be due to the fine material plugging the drainage ple the lowest. The order for the components in the furnish
channels as the mat is formed [9].The viscous resistance coef- with decreasing drainage resistance is the stone groundwood,
ficient (a) should be proportional to the porosity of the mat followed by the TMP, broke, BCTMP and finally the refined

VOL. 1: NO. 6 TAPPI JOURNAL 17

PAPERMAKING

softwood kraft pulp, although the latter was refined very little. 10. Hung, L., Leung, W. K., and Green, S. I., "Pulp fiber mat
The headbox sample had a higher drainage resistance coeffi- permeability measurement under simulated forming
cient than any of the component [Link] believe this resulted conditions," 2000 TAPPI Engineering Conference
from the introduction of fine particles contained in the white Proceedings, TAPPI Press, Atlanta, Georgia, USA (2000).
water used to dilute the component pulps to the headbox con- 11. Arslan, K., Bousfield, D.W., and Genco, J.M., TAPPI J.
80(1): 254(1997).
[Link] important point is that the drainage properties of
12. Paradis, M.A., Genco, J.M., Hassler, J.C., and Wildfong,
these various pulps can be characterized by the drainage resist- V.J., “Determination of drainage resistance coefficients
ance test and inserted into the available computational models. under conditions of known sheer rate”, 2001 TAPPI
The new aspect reported here is that the shear rate and shear Engineering Conference.
stress for these drainage conditions are known. Figure 10 also 13. Bird, R.B., Stewart, W.E., and Lightfoot, E.N., Transport
serves as the start of a database that can be developed for vari- Phenomena, p. 200, Wiley Press, New York., New York,
ous pulps and can be used for designing formers. USA (1960).
14. Franzen, M. F., "Simulation and optimization of the
CONCLUSIONS papermaking sheet forming process," Proceedings of
the 2000 ASME Fluids Engineering Division Summer
The flow resistance data for the forming fabric agree with
Meeting, ASME International, New York, New York, USA.
the Ergun [Link], if the equivalent diameter, thick-
ness, and the void fraction of the fabric are known, the flow Received: July 19, 2001
resistance of the wire can be [Link] relationship may be Accepted: December 20, 2001
of use in designing fabric for papermaking applications.
The modified drainage apparatus described here permits This paper was originally presented at the 2001 TAPPI
Engineering Conference.
drainage resistance coefficients to be determined under a
known shear condition at high drainage rates and at short This paper is also published on TAPPI’s web site
[Link] low shear rates, the drainage resistance increases with ([Link]) and summarized in the August Solutions! for
increasing shear rate. At higher shear rates, the resistance People, Processes and Paper magazine (Vol. 85 No. 8).
decreases with increasing shear rate. This decrease may be a
result of the mat being removed by shear forces from the wire.
The drainage resistance coefficients for both the pulp and
headbox samples are a function of the basis weight under shear Paradis, Genco, Bousfield, and Hassler are with the Department
[Link] increase in drainage resistance coefficient with of Chemical Engineering, University of Maine, Orono, Maine,
basis weight may be related to pore plugging which alters the USA; Wildfong is with Johnson Foils Corp., Springfield,
effective void fraction of the forming mat. TJ Massachusetts, USA. Email Paradis at mparadis@[Link].

LITERATURE CITED
1. Zhao, R.H., and Kerekes, R.J., J. Pulp Paper Sci. 21(3):
97(1995).
2. Zahrai, S., and Bark, F.H., Nordic Pulp Paper Res. J.,
10(4): 245(1995).
3. Roshanzamir A., Green, S.I. , and Kerekes, R.J., J. Pulp
Paper Sci. 24(11): 364(1998).
4. Green, S.I., J. Pulp Paper Sci. 23(7): 353(1997).
5. Green, S.I., Zhao, R.H., and Kerkes, R.J., J. Pulp Paper Paradis Genco Bousfield
Sci., 24(2): 60(1998).
6. Dalpke, B., Green, S., and Kerkes, R.J., “Modelling of jet
impingement in twin-wire paper machines: impinge-
ment on one fabric,”TAPPI J., 84(2):44 (2001); published
online at [Link].
7. Mantar, E., Co, A., and Genco, J.M., J. Pulp Paper Sci.
21(2): 44(1995).
8. Wildfong, V.J., Genco, J.M., Shands, J.A., and Bousfield,
D.W., J. Pulp Paper Sci. 26(7): 250(2000).
9. Wildfong, V.J., Genco, J.M., Shands, J.A., and Bousfield,
D.W., J. Pulp Paper Sci. 26(8): 280(2000). Hassler Wildfong

18 TAPPI JOURNAL AUGUST 2002