1.

TITLE
Filtration Unit

2. INTRODUCTION
Filtration is a process that separates liquid from a suspension with the use of porous
medium (Coulson & Richardson, 2002). It is accomplished by removing solid particles from a
fluid by passing the a fluid through a medium. The suspension is forced through the membrane or
medium by pressure or force. The important elements that are involved in a filtration process in
the slurry, the medium, the filtrate and the cake. The suspension which is separated through a
medium is called the slurry. The medium is porous and allows for selective passage of the particles
that make up the slurry. The objective of filtration is to separate solid particles from the fluid in
which it is present in.
At the industrial level, filtration ranges from simple straining to complex separation
processes. The feed in industries can be gases or liquids. Usually, the feed is subjected to pre-
treatment to increase the rate of filtration by heating, recrystallizing or adding filter aids. Some
application of filtration in the industry is in ethanol production, mining, water treatment, steel mill
filtration, oil and gas desanding and removal of crumbs from hot cooking oil among many other.
Filtration is an important part of the production line of any process to increase the yield of the
desired product and decrease impurities.

3. LITERATURE REVIEW
 Filtration is a widely used unit operation of Chemical Engineering Processes which
separates solids from fluids. According to Coulson & Richardson Volume 2, filtration is the
separation of solids form a suspension in a liquid by means of a porous medium or screen which
retains the solids and allows the liquid to pass in termed filtration. This means that fluids or slurries
are forced through the pores of the medium which causes solid particles larger than the pores of
the filter medium to be removed. As time passes, particles deposited will slowly build up over time
and this would lead to the formation of layer of solids which is called a filter cake. As a thicker
filter cake is formed, flow resistance increases and this causes pressure to increase which greatly
affects the rate of filtration. It is applicable in various industries such as wastewater treatment,
paper and pulp industry and as well as food and beverages industry.
In wastewater treatment industries, it is a must to make sewage harmless before it is
disposed to prevent pollution. One of the most important reasons to treat wastewater is to control
the Biochemical oxygen demand (BOD) of the water, which is the oxygen required for biological
decomposition of biodegradable organic matter under aerobic conditions. Filtration is one of the
process used in order to treat wastewater. It is stated in Basic Environment Engineering by R.C.
Gaur that filtration is a physical and chemical process for separating suspended and colloidal
impurities from water by passage through a porous bed made up of gravel and sand.
There are a lot of different types of filter equipments. Filter equipments consists of bed
filters, bag filters, filter press, pressure leaf filters, cartridge filters, vacuum filters, band filters,
rotary drum filters and rotary disc filters. The experiment that is about to be conducted is based on
a plate and frame filter press which is one type of filter press.
In general, plate and frame filter press functions as solids removal from slurries or fluid. It
has a series of plates and frames in alternating arrangements which are covered with fiber cloths
or filter papers. In order to remove solids from slurries or fluids, selection of a filter paper or fiber
cloths are of high importance.
 Figure 1- Filtration of slurry in filter press
According to Tarleton and Wakeman (2007), they state that, the term filter medium can be
defined as ‘any material that, under the operating conditions of the filter, is permeable to one or
more components of a mixture, solution or suspension, and is impermeable to the remaining
components’ (Purchas and Sutherland, 2002). Tarleton and Wakeman (2007) also claimed that,
selection of the filter medium must take account into many factors; criteria by which a medium is
assessed include the permeability of clean medium, its particle retention capability and the
permeability. So, if selection of filter medium is not critically considered before operation filter
presses, there might be occurrence of using filter mediums with too big pores which will eventually
end up to a very small cake formed, or no cake formed at all. In another source (McCabe et. al,
2005) states that the septum in any filter must meet the following requirements:

1. It must retain the solids to be filtered, giving a reasonable clear filtrate.

2. It must not plug or blind.
3. It must be resistant chemically and strong enough physically to withstand the
process conditions.
4. It must permit the cake formed to discharge cleanly and completely
5. It must not be prohibitively expensive.

Canvas cloths are common filter mediums but when corrosive liquids are used, it requires the use
of other filter media, such as woolen cloth, metal cloth of monel or stainless steel, glass cloth, or
paper.

PLATE AND FRAME FILTER PRESS SYSTEM

 Figure 2- Typical process flow of filter press

This type of filter allows pressurized filtration of a slurry mixture to remove solids. A set
of filter plates is sandwiched together in series with a configuration similar to the plates on a plate-
and-frame heat exchanger. After the plates are compressed together hydraulically, a channel
throughout all the plates allows the slurry to be pumped into one side of each of the plates. The
filtrate goes through the filter media on the plate leaving the solid cake on the media. Filtrate on
the other side of the filter plate enters a channel, joins the filtrate coming off the other plates in the
filter, and leaves the equipment. When the filter plates are filled with solids, the solids can be
washed and the press opened to discharge the solids off each plate. The filter press consists of a
head and follower that contain in between a pack of vertical square plates that are supported by
side or overhead beams. The head serves as a fixed end to which the feed and filtrate pipes are
connected and the follower moves along the beams and presses the plates together during the
filtration cycle by a hydraulic or mechanical mechanism.

Each plate is dressed with filter cloth on both sides and, once pressed together, they form
a series of chambers that depend on the number of plates. The plates have generally a centered
feed port that passes through the entire length of the filter press so that all the chambers of the plate
pack are connected together. Likewise, four corner ports connect all the plates and collect the
mother and wash filtrates in a "closed discharge" towards outlets that are located on the same side
as the feed inlet. Some filter presses have plates that are fitted with cocks at their lower side so that
the filtrate flows in an "open discharge" to a trough and serve as "tell tales" on the condition of the
filter cloth by the clarity of the filtrate that passes through each chamber. The disadvantage of this
arrangement is that it cannot be used with filtrates that are toxic, flammable or volatile. A typical
flow scheme is illustrated in Figure 2 .

Often special measures are taken to ease cake discharge and enhance filtration. The
measures taken include precoating and the addition of body aid. Filterpresses were introduced
nearly 100 years ago and have been applied extensively in dewatering waste sludges. They were
considered labor intensive machines hence they did not find much acceptance in the sophisticated
and highly automated process industries. It was not until sometime in the 1960's that this image
changed by the introduction of advanced mechanisms that were oriented towards obtaining low
moisture cakes that discharge automatically and enable the washing of the cloth at the end of the
filtration cycle. When the contaminants are gelatinous and sticky it forms a barrier that avoids
cloth blinding. Likewise the interface between the precoat and the cloth departs readily so the cake
discharges leaving a clean cloth. When a clear filtrate is required immediately after the filtration
cycle commences otherwise recirculation must be employed until a clear filtrate is obtained.

Main features of a modern-day filter press include shuttle shifters that separate the plates
one by one for cake discharge at a rate of 5-6 seconds per plate. A special design of the shifting
mechanism ensures that two adjacent plates are not pulled together due to sticky cakes. Filter
presses are built for operating pressures of 7, 10 and 15 bar for cake squeezing and the largest
available plates are 2 by 2 meters so the hydraulic pressure system that holds the closing force of
the plates is designed accordingly. Filter press plates are available in various materials of
construction such as cast iron, aluminum alloys, and plastics such as high-density polypropylene.

The major area of development, apart from automation, was in the design of the plates
since thermoplastics have enabled new structural concepts which were not possible with metallic
plates. Plastic composite plates have the following unique features:
 ● Lower plate weight has reduced the downtime for shuttle shifting during the cake
discharge mode.
● Effective filtration area has gone up since with the largest available plates of 2 by
2 meters, having a 20 mm recess and 150 chambers, the area is about 1000 m2 with
a cake capacity of 20 m3

Filter presses are used in batch processes. They are simple to operate and flexible in
operations. They can be used at high pressures,when necessary, if viscous solution is being used
or the filter cake has a high resistance.
Medium flow resistance, Rs , is an important parameter to indicate the flux through the
medium in filter press. The medium flow resistance,Rs, is calculated using correlation where
volume of cumulative filtrate over time is first measured and a straight line is plotted based on
Equation 1,
𝑡
= aV+b Equation 1
𝑉
Where,
t= cumulative time (s)
V= volume of filtrate (L)
a= slope of the graph
b= y-intercept of the graph

After that, the medium resistance flow, Rs can be determine using Equation 2,

𝑏𝛺𝛥𝑃
Rs= Equation 2
𝜂
Where,
b= intercept point of the graph from Equation 1
𝛺= surface area of the filter (m2)
 𝛥𝑃=pressure difference between feeding pressure and downstream
pressure (bar)
𝜂=dynamic viscosity of filtrate at 25ºC (= 0.89 x 10-3 Pa.s)

Figure 3- Schematic diagram of the pilot plant setup

4. METHODOLOGY
Part 1: Preparation of solution
1. V1 and V9 were closed.
2. Tank 1 was filled with 150 L of tap water.
3. 3 kg of filter aid was weighed.
 4. V4 was opened. Then, the air compressor switch was turned on. After that, V11
was partially opened until the pressure reading for air is 1 bar.
5. 3 kg of filter aid was slowly poured into Tank 1. Do not turn off the air compressor
until the end of the experiment.

Part 2: Placing of filtration plates (3 plates)

1. The main switch at the equipment switchboard was turned on.
2. 3 plates were placed in between the spacers. The 2 end filtration plates are the ones
with flat surface.
3. 2 frames were placed in between the plates. The position of the holes of the frames
is ensured match those of the plates.
4. The mobile spacer was pushed until it pressed the filtration plate.
5. The long cylinder bar was placed in horizontal position inside the filter.
6. V5 was ensured closed.
7. The PVC filter cake was ensured placed beneath the filtration plates.

Part 3: Plates pressurization

1. The vent valve of the oil pump body was closed.
2. The valve on the cylinder body was opened.
3. The cylinder was manually pumped in quick motion until pressure of 150 bars is .
4. The valve on the cylinder body was closed to maintain the oil pressure in the
cylinder body.
5. The vent valve on the oil pump body was opened.

Part 4: Filtration using 3 plates system

1. V2 and V5 were ensured closed.
2. V1 and V3 were ensured opened.
3. The triangle sign at V7 was ensured connects the pipe from Tank 1 to the pump
direction.
4. At the control box, the filtration pressure, 𝛥𝑃was set to 0.2 bar.
5. V2 and V8 were ensured closed.
 6. The “Filtration on” button at the control box was pressed.
7. The suspension circulated for 5 minutes in order to obtain homogeneity in filter
feeding line.
8. 2 beakers and 1 timer were used for data recording.
9. The small tube from V6 was placed inside a beaker as preparation for filtrate
volume measurement.
10. V6 was ensured closed. Then, V2 was opened.
11. After V2 was fully opened, V3 was closed.
12. V6 was opened.

Part 5: Data collection for 3 plates filtration

1. The timer was started once the filtrate enters the beaker.
2. The filtrate volume from V6 is measured every 1 minute and the data was recorded.
3. Measuring was stopped after 10 minutes.
4. The “Filtration off” at the control box was pressed.
5. V2 was closed and V3 was fully opened.

Part 6: Filtration plate depressurization

1. The valve at the oil pump body was closed.
2. The cylinder was pumped manually until it reaches 200 bar.
3. The valve ate the cylinder body was opened.
4. The valve at the oil pump body was opened.
5. The position of the cylinder was waited until its back to initial position.
6. The long cylinder bar was taken out.

Part 7: Preparation for 5 plates filtration

1. The cake from the filtration plates was removed using spoon/spatula.
2. The filtration clothes was cleaned with water.
3. 5 plates were placed between the plates. The 2 end filtration plates are the ones with
flat surfaces.
 4. 4 frames were placed in between the plates. The position of the holes of the frames
are ensured match those at the plates.
5. Step 4-2 in Part 2 were repeated but the short cylinder bar was used.
6. Step 1-5 in Part 3 were repeated.

Part 8: Filtration using 5 plates system

1. All the steps in Part 4 were repeated.
2. All the steps in Part 5 were repeated.
3. All the steps in Part 6 were repeated.

Part 9: Rinsing of pilot plant

1. The remaining suspension in Tank 1 was drained by closing V1 and opening V9.
2. When Tank 1 is empty, V1 was closed and the power switch of air compressor was
turned off.
3. V1 and V2 were closed.
4. Tank 1 is rinsed using tap water until the filter aid is cleared.
5. The power at the main switchboard is turned off.

5. RESULTS AND DISCUSSION

5.1 Results
Medium flow resistance, 𝑅𝑠 can be calculated using correlation where volume of
cumulative filtrate over time is first measured and a straight line is plotted based on;
𝑡
= 𝑎𝑉 + 𝑏
𝑣
Where,
t : cumulative time(s)
V : volume of filtrate (L)
a : slope of the graph
b : intercept of the graph
 After that, the medium resistance flow, 𝑅𝑠 can be determined using the equation
below;
𝑏𝛺𝛥𝑃
𝑅𝑠 =
𝜂

Where,
B : intercept point of the graph
𝛺: surface area of filter (= 1𝑚2 )
𝛥𝑃: pressure difference between feeding pressure and downstream pressure(= 0.2bar)
−3
𝜂 : dynamic viscosity of filtrate at 25∘ 𝐶(=0.89× 10 Pa.s)

conditions : 3 plates.
𝛥𝑃 = 0.2 𝑏𝑎𝑟
Table 1: Results for 3 plates

Volume of Volume of
𝑡
Time(s) filtrate,V(L) Filtrate,V(𝑚3 ) × 10−3 (𝑠. 𝑚−3 )
𝑣

0 0.000 0.00000 0.0000000

60 6.500 0.00650 9.2307692

120 3.500 0.00350 34.2857143

180 3.150 0.00315 57.1428571

240 3.500 0.00350 68.5714286

 300 3.500 0.00350 85.7142857

360 3.625 0.00363 99.3103448

420 3.325 0.00333 126.3157895

Figure 4: the graph of t/v versus V for 3 plates.

Based on Figure 4, the intercept b, is 51.798. The value for medium resistance, Rs
can be calculated using Equation 2.
𝑏𝛺𝛥𝑃
𝑅𝑠 =
𝜂
𝑠 𝑁
51.798( 3 ) × (1𝑚2 ) × (0.2 × 105 2 )
𝑅𝑠 = 𝑚 𝑚
−3 𝑁
0.89 × 10 .𝑠
𝑚2
𝑅𝑠 = 1.164 × 109 𝑚−1

conditions : 5 plates
𝛥𝑃 = 0.2 𝑏𝑎𝑟
Table 2 : Results for 5 plates
 Volume of 𝑡
Time(s) filtrate,V(L) Volume of Filtrate,V(𝑚3 ) × 10−3 (𝑠. 𝑚−3 )
𝑣

0 0.000 0.00000 0.0000000

60 9.000 0.00900 6.6666667

120 8.000 0.00800 15.0000000

180 6.625 0.00663 27.1698113

240 4.400 0.00440 54.5454545

300 3.600 0.00360 83.3333333

360 3.500 0.00350 102.8571429

420 3.500 0.00350 120.0000000

480 3.550 0.00355 135.2112676

540 3.400 0.00340 158.8235294

600 3.600 0.00360 166.6666667

Figure 5: tha graph of t/v vs V for 5 plates

Based on Figure 5, the intercept b, is 107.51 . The value for medium resistance,
𝑅𝑠
can be calculated using equation 2.
𝑏𝛺𝛥𝑃
𝑅𝑠 =
𝜂
𝑠 𝑁
(107.51 3 ) × (1𝑚2) × (0.2 × 105 2 )
𝑅𝑠 = 𝑚 𝑚
−3 𝑁
0.89 × 10 .𝑠
𝑚2
𝑅𝑠 = 2.416 × 109 𝑚−1

5.2 Discussions

The purpose of this experiment is to study the effect of number of plates on medium
resistance in filter press system for solid-liquid separation.The experiment started with
weighing 3 kg of filter aid and dissolved it into 150 litre of tap water in slurry tank. During
the experiment, filtrate volume was measured every 1 minute for 3 plates filtration until
there were no filtrate came out from the equipment. The steps were repeated with 5 plates
filtration. The performance of the filter and number of plates can be observed and
calculated from the result obtained and the graph plotted for both of the plates used.

For 3 plates filtration, the time taken until there were no filtrate came out from the
equipment was 420 seconds. Based on the result taken in Table 1, Figure 4 were plotted to
find medium resistance flows. From Figure 4, the equation of the line is y=2442.4X +
51.798 giving the interception b=51.798. By substituting the value b into equation 2
medium resistance flow,𝑅𝑠 can be calculated which is equal to 1.164 × 109 𝑚−1 .

Whereas for table 2 shows the result for 5 plates filtration. There was no filtrate
came out after 600 seconds. Based on Figure 5, the equation of the line is y=-8150.7X +
107.51 giving the value of the interception b= 107.51. The calculated medium resistance
flow,𝑅𝑠 using equation 2 is 2.416 × 109 𝑚−1 .

Based on the tabulated data in table 1 and 2, the first volume taken for both of the
plates used was the highest. This is due to no filter cake formed causing the pressure to its
highest since there is still less resistance. The volume decreases with time because filter
cake is starting to form, and more pores are block, and less filtrate can get through. For 3
plates filtration, shorter time needed for filtrate to stop coming out from the equipment
compared to 5 plates filtration as 3 plates filtration stop at 420 seconds and 5 plates
filtration stop at 600 seconds giving the differences of 180 seconds. This is because as the
number of plates increases, the area for filtration increases as there are more available open
pores for filtrate to go through. However, the longer the experiment were to run, more open
pores will be blocked causing the medium resistance flow,𝑅𝑠 to increase.

There are three potential contributors of error to the medium resistance flow
determination which is first the experimenter does not check the equipment before
conducting the experiment especially when handling an instrument which consist of more
than 5 valves and a pump. To overcome this problem, pre read lab manual must be done
and all valves must be checked to ensure that all of them are close prior to conducting the
experiment.

Second potential contributor of error to the medium resistance flow determination

is the plates and frames assembled alternatively with fibre cloth over each side of plates
does not tightly secured using the mobile spacer. This will cause filtrate to come out from
the plates and frames and causes leaking. To avoid this error, mobile spacer must be push
until it presses the filtration plate tightly. The position of the holes of the frames must also
match those at the plate.

The third potential contributor of error to the medium resistance flow determination
is parallax error. Parallax error will lead to deviation in the accuracy of the data obtained.
The water is highly turbulent and for the first volume taken for both of the plates used, the
value must be the highest due to no filter cake formed causing the pressure to its highest
since there is still less resistance. It will be harder to record the reading for the first two
data since there are more volume to measure and the data must be recorded every 1 minute
interval. Therefore, point estimation must be used.
 Table 3: Identification and function of each valve:

V1 Channel water from the main tank containing known

concentration of filter aid to filtrate rinsing water

V2 Channel the main tank to filtration chamber

V3 Send water containing filter aid back to the main tank if

there is blockage in filtration chamber

V4 Contain pressure meter that monitor the hydraulic press

pressure.

V5 (vent Lock the hydraulic press arm. If this valve is open, any
valve) movement on hydraulic arm will not affect the pressure.

V6 Connect hydraulic press arm with its cylinder body.

(cylinder Opening this valve will depressurized the arm and locking
valve) the pressure to the plate.

V7 Contain triangle sign that connects pump to from tank 1

V8 Contain ball valve to discharge water from the tube in order

to clean the unit

V9 Drain the main tank. It is only open after the experiment to

clean the main tank

6. CONCLUSION AND RECOMMENDATION

After the experiment, we were able to achieve our objective which is to study the
effect of number of plates on the medium flow resistance in filter press system for solid-
liquid separation. We obtained separated filtered water and mud cake at the end of
experiment.
The effect of number of plates on the medium flow resistance in filter press system
for solid-liquid separation was determined by plotting graph t/v versus V for 3 plates and
 5 plates. By substituting intercept value into the Equation 2 , we obtained the value for
medium resistance for 3 plates and 5 plates. 5 plates have higher medium resistance
compared to 3 plates as more number of plates was used in the filter system. More number
of plates means larger area available for filtration and more filtrate can go through to the
available pores. Thus, more volume of filtered water can be obtained. We can conclude
that the experiment was successfully conducted since we were able to the study the effect
of number of plates on the medium flow resistance in filter press system for solid-liquid
separation
Apart from that, there are some recommendation for this experiment, in order to
make sure that there are no errors and mistake happened during the experiment. Errors can
cause the data deviate from the theoretical value. In this experiment, errors might happened
during reading volume of the filtered water in the beaker. Eyes must be parallel to the
measurement even though we need to take reading within 1 minute.
Other than that, student should follow the guidelines and the rules before doing the
experiment. They need to wear lab coats, gloves and wear shoes for safety purposes. They
also need to make sure they are alert with the precautions mentions in the lab manual and
never play around during the experiment being conducted.

7. REFERENCES

1. Cake Filtration Equipment - Drinking Water - Climate Policy Watcher. (2018).Climate-

policy-watcher.org. Retrieved 18 March 2018, from https://www.climate-policy-
watcher.org/drinking-water/cake-filtration-equipment.html
2. Bacon, D., Towler, G., Sinnott, R., & Lyons, W. (2018). Filtration Laboratory Report |
Filtration | Chemical Engineering. Scribd. Retrieved 18 March 2018, from
https://www.scribd.com/doc/285720796/Filtration-Laboratory-Report
3. Coulson, J., & Richardson, J. (2002). Coulson and Richardson's Chemical Engineering
(Vol. 2). Woburn,MA: Elsevier Science.