Flow Meters By FPG

Customer Service Available

Most Sophisticated And Flexible Flow

Meters On The Market. Buy Now!

fpg.co

OPEN
Distillation columns or systems are designed some thing in the same way as
a sequence of individual stills as previously mentioned. Each ‘still’ area
includes a variety of ‘TRAYS’ or getting in touch with gadgets organized top
to bottom above one another in the pillar. These containers or contactors
carry fluid and fumes into romantic get in touch with in order to acquire
the needed separating of the combination. The size of the structure and the
variety of containers or getting in touch with gadgets it contains, relies on
the cleanliness of the ‘Fractions’ needed.

be of the following types:
Columns for the distillation procedure can

The ‘PACKED’ Tower

The ‘TRAY’ Tower

1. The Packed Towers

As its name indicates, the packed structure is a straight, metal pillar which
contains ‘Beds’ of packaging content which are used to carry the increasing
vapours into romantic get in touch with with dropping fluid within the
structure. The warm included to the combination before coming into the
structure partly vaporises the combination and the vapours increase up the
structure and begin to awesome.

The fluid drops towards the end of the structure. At the structure base, in
common, more warm is included to the fluid by a ‘Reboiler’ which may be
vapor warmed or a petrol shot heater kind.

The inclusion of warm here causes more vapours to increase up the pillar.
As the two stages of the combination – dropping fluid and increasing fumes
– come together, light elements are removed out of the fluid and get into the
gas stage while hefty elements in the fumes are compacted into the fluid
stage.

In this way, as the fumes increases and progressively calme down, it

becomes less hefty and, as the fluid drops, it becomes warmer and bulkier.

With this kind of distillation pillar there is usually only a top and base
product. The quality of the products relies on the size of the structure, the
variety of getting in touch with gadgets, the structure heat range and stress
and their control, and the speed of the increasing vapours.

The kind of packaging components used, also results in the separating

procedure. The packaging can be of such kinds as:

Ceramic Raschig Jewelry, Stainless-steel Steel Pall Jewelry or Clay Horse

saddles etc. See Figure: 10.

2. THE TRAY TYPE TOWER

The kinds of tray used in distillation content are as follows:

THE SIEVE TRAY is a steel plate containing drilled gaps through which the
increasing fumes can complete into the liquid streaming across the tray.
Figure: 11

TRAY is just like the sieve type but, each opening is fixed with a
THE VALVE
flapper valve which starts as fumes goes through the opening. This kind is
used where fumes speed is not continuous and the valves avoid liquid from
throwing through the gaps at periods of low gas speed. Figure: 12
THE BUBBLE-CAP TRAY is the most effective separating program but,
is also
the most expensive. It includes a variety of ‘Chimneys’ or ‘Risers’ (small,
brief tube joints set into the tray), through which the fumes can complete.
Fitted over the riser is a ‘Cap’ which causes the increasing fumes to convert
through 180 °. This causes the gas to ‘Bubble’ through the liquid streaming
across the tray. The liquid stage on the tray is managed below the top of the
riser to avoid throwing of liquid down the structure. Figure: 13
Each of the above containers also has a ‘WEIR’ that preserves the
liquid
stage on the tray. As the liquid moves over the weir, it goes into a
‘DOWNCOMER’ – (a brief pipe), that provides the liquid down to the tray
below. The downcomer store is below the outer lining area of the liquid on
the tray below, performing as a closure to avoid gas from skipping the tray
above.

November 1st, 2012 in Distillation Technology

DISTILLATION COLUMN CONTROL
http://seperationtechnology.com/distillation-column-control/

Trillium Flow Technologies

Outstanding Flow Control
Our Products Are Backed By Expertise &

Support. Contact Us Today To Learn More.

trilliumflow.com

OPEN

1. REFLUX RATIO
The reflux to a structure top is used to management the top heat range
thereby manages the cleanliness of the expense item.

The quantity of reflux compared to the item is known as the ‘Reflux Ratio’.

An example of this is as follows:

The expense fluid from a distillation column is separated into 4 m3 hourly

reflux and 2 m3 hourly item. Therefore:

When reflux amount is improved, the quality of reflux improves. Gerd

symbolizes chilled, compacted top item came back to the structure top and,
as such it is being reprocessed. The top item will therefore be more pure. In
general, the higher the reflux amount, the less the number of containers
needed for a given separating.
However, too great a amount may cause surging in the structure leading to
inadequate separating and causing ‘off-spec’ items throughout the program.
The reflux amount is normally managed by a heat range operator in the
fumes store which functions a management device in the reflux push
release. An increase in structure top heat range will cause the device to
start, improving the reflux amount, and viceversa.

2. TEMPERATURE GRADIENT
Temperature (and pressure) management of a distillation structure will
regulate the cleanliness of the items. The management of top heat range is
as mentioned above in ‘reflux’. Control over the supply inlt heat range and
that of the reboiler are also very essential.

Again, if supply and bottom temperature ranges are too great, too much
hefty fumes will rise up the structure and put side-stream items off-spec.
This condition, along with great reflux amount will again cause to surging
and inadequate separating. Opposite conditions can cause to fluid hunger
across the containers and again, a very disappointed process will outcome.

The cautious management of top heat range, supply and reboiler

temperature ranges, together with stress management, will provide the
preferred heat range information across the structure.

Remember, changes in stress will impact the steaming points of the

elements in the raw oil. The fumes demands therefore, will also be affected
and again, if the management factors are wrong, the program will be
ineffective.

Examples:

range will outcome in hefty elements in the expense item.
High top heat
Low top heat range will outcome in a less hefty top item.

High supply heat range will provide bulkier side-streams

and vice-versa.
High reboiler heat range will produce bulkier soles item and pass bulkier

vapours up the structure to impact the side-streams.

elements in the
Increased stress in the program will provide less hefty
expense LIQUID item and reduce its Initial Boiling Factor, whereas the FBP
is managed by the structure top heat range.
It can be seen that cautious, precise management of the factors is very
essential to experience the needed qc of the items.

Also, with respect to the cleanliness of the side-streams, management of the

burning towers’ vapor supply is very essential.

A further point is, that high-water content in the raw supply will cause
stress deluges as the water vaporises in the structure. The raw oil should be
as water free as possible.

Many modern distillation models are managed under great machine. This
method, due to the machine reducing the BP’s of the elements of the
combination to be separated, also decreases the quality of heat energy
needed to vaporise the elements.

VACUUM DISTILLATION TOWER OVERHEADS SYSTEM

Figure: 15
In the above plan, the outer lining place condenser is a ‘Total Condensing’
unit. This means that all liquids that can be compacted are modified to
fluid. Due to this, a machine is established in the structure the stage of
which relies on the degree of moisture build-up or condensation permitted
to take place. This is managed by the stage of distillate and how much of the
condensing surface place is protected. The fluid stage and therefore the
quality of condensing surface place available will decide the stage of
machine (Absolute Pressure) of the program.

The PRC is therefore managing the available condensing place on the

chilling pipes. Increasing overall stress (decreasing vacuum), will start the
management device, the fluid stage in the condenser will fall thus
introducing more condensing place to the fumes. More fumes will reduce
and therfore the stress will fall returning again – and vice-versa.

A small quantity of uncondensible fumes will tend to develop up in the

structure and the outer lining place condenser which, if permitted to
develop up, will gradually eliminate the machine.

The ejector is used to eliminate the uncondensibles which are approved

into the separator after moving through the ejector condenser which
condenses the ejector vapor – thus helping to maintain the machine, while
the uncondensed fumes are fed via a check-valve or management device to
environment or width program. The barometric closure cycle maintains a
head of fluid which will prevent the machine taking fumes returning out of
the separator.

September 15th, 2012 in Operations of Distillation System

 Bioanalytics

ITSI-Biosciences Open

The boiling point of A is that at which the mole fraction of A is 1. The

boiling point of B is that at which the mole fraction of A is 0. In this
example, A is the more volatile component and therefore has a lower
boiling point than B. The upper curve in the diagram is called the dew-point
curve while the lower one is called the bubble-point curve.

The dew-point is the temperature at which the saturated vapour starts to

condense.

The bubble-point is the temperature at which the liquid starts to boil.

The region above the dew-point curve shows the equilibrium composition
of the superheated vapour while the region below the bubble-point curve
shows the equilibrium composition of the subcooled liquid.

For example, when a subcooled liquid with mole fraction of A=0.4 (point A)
is heated, its concentration remains constant until it reaches the bubble-
point (point B), when it starts to boil. The vapours evolved during the
boiling has the equilibrium composition given by point C, approximately
0.8 mole fraction A. This is approximately 50% richer in A than the original
liquid.

This difference between liquid and vapour compositions is the basis for
distillation operations.

Relative Volatility
yi = mole fraction of component ‘i’ in the vapour

xi = mole fraction of component ‘i’ in the liquid

Thus if the relative volatility between 2 components is very close to one, it

is an indication that they have very similar vapour pressure characteristics.
This means that they have very similar boiling points and therefore, it will
be difficult to separate the two components via distillation.

VAPOUR LIQUID EQUILIBRIA

Distillation columns are designed based on the boiling point properties of

the components in the mixtures being separated. Thus the sizes,
particularly the height, of distillation columns are determined by the
vapour liquid equilibrium (VLE) data for the mixtures.

Vapour-Liquid-Equilibrium (VLE) Curves

This particular VLE plot shows a binary mixture that has a uniform vapour-
liquid equilibrium that is relatively easy to separate. The next two VLE plots
below on the other hand, shows non-ideal systems which will present more
difficult separations. We can tell from the shapes of the curves and this will
be explained further later on.
The most intriguing VLE curves are generated by azeotropic systems. An
azeotrope is a liquid mixture which when vaporised, produces the same
composition as the liquid. The two VLE plots below, show two different
azeotropic systems, one with a minimum boiling point and one with a
maximum boiling point. In both plots, the equilibrium curves cross the
diagonal lines, and this are azeotropic points where the azeotropes occur.
In other words azeotropic systems give rise to VLE plots where the
equilibrium curves crosses the diagonals.
Note the shapes of the respective equilibrium lines in relation to the
diagonal lines that bisect the VLE plots. Both plots are however, obtained
from homogenous azeotropic systems. An azeotrope that contains one
liquid phase in contact with vapour is called a homogenous azeotrope. A
homogenous azeotrope cannot be separated by conventional distillation.
However, vacumn distillation may be used as the lower pressures can shift
the azeotropic point.Alternatively, an additional substance may added to
shift the azeotropic point to a more ‘favourable’ position.

When this additional component appears in appreciable amounts at the

top of the column, the operation is called azeotropic distillation.

When the additional component appears mostly at the bottom of the

column, the operation is called extractive distillation.
They may be separated in 2 distillation columns since these substances
usually form two liquid phases with widely differing compositions. The
phases may be separated using settling tanks under appropriate conditions.

January 10th, 2012 in Vapor Liquid Equilibria

Multimode Reader
Convenient software options. Ergonomic design. Get

a quote now.
Berthold Technologies Open
The hot supply now goes into the structure into the ‘Flash Zone’. At this
point, due to the greatly increased volume of the pillar, the less hefty
elements of the oil ‘ Display Off ‘ (vaporise), and increase up the pillar. The
hot fluid will fall towards the pillar base.

The base area of the pillar, below the Display Area, known as the ‘Stripping
Section’, contains containers – usually Bubble-cap or Filter type. The
structure base fluid is re-circulated & re-heated in a vapor or shot ‘Reboiler’
which pushes off vapours of mild finishes and some of the hefty finishes in
the fluid. These vapours increase up-wards through the containers and
contact the down-flowing fluid. This activity further eliminates (strips out),
mild finishes from the fluid.

The top area of the structure, above the flash zone, is known as the
‘Rectifying Section’. Here again, the increasing fumes passing through the
containers, connections the fluid streaming across them.

Action of the Trays Each plate in the structure is performing like a single
still as discussed in ‘Batch Distillation’. As we increase above the flash zone,
each following plate is a little bit chilly than the plate below.

The down-flowing fluid, as it goes across the containers is becoming

warmer and bulkier as mild finishes steam off into the fumes stage. On the
other hand, the increasing fumes is becoming chilly and less hefty as
bulkier finishes reduce into the fluid on the plate.

The down-flowing fluid is known as ‘Internal Reflux’ and works in the same
way as in Figure: 7 (Page 14), where the fluid is came back to the previous
still.
The least heavy elements of the oil combination leave the top of the
structure as fumes. This is fed through condensers – usually water-cooled -
and the condensate, usually Naphtha and water, goes into the ‘Overhead
Device or Accumulator’.
In the recipient, mild fumes also build up. The management of these fumes,
(to a petrol program or flare), also manages the stress on the distillation
procedure at the needed stage.

The Naphtha fluid forms an program above the water. The water is cleared
away under management, to disposal. The Naphtha, also under stage
management, is separated into two – some is came back to the structure top
plate as ‘External Reflux’ which is used to management the structure top
warm range and thereby help to management the naphtha top quality.

The staying naphtha from the recipient is piped to storage space and / or to
other processes. The items making the part of the pillar -called ‘Side-
streams’, are usually approved through ‘Stripping Towers’ where an
hypodermic injection of superheated vapor eliminates last records of mild
finishes to fulfill the requirements needed for the item. The mild finishes
and vapor are approved back into the structure.
The side-stream items pass from the burning systems through supply / item
exchangers and water chillers to storage space.
The structure base item as already mentioned, is reheated in a reboiler to
remove mild finishes and to provide burning fumes in the structure. The
ultimate base item, such as hefty petrol oil is treated away via feed/product
warm exchangers and water chillers to storage space. In some distillation
systems, superheated vapor may be treated into the structure base to assist
the burning procedure.

Crude oil distillation is often carried out under machine circumstances. The
machine is produced by taking the expense fumes from the structure by
vapor ejectors via surface condensers.

The explanation of oil separating given above, is that of a primary program.

Raw oils also produce substances, waxes, gasolines, lubrication and many
other items in everyday use.

October 5th, 2012 in Operations of Distillation System

EVAPORATION and VAPOUR PRESSURE
http://seperationtechnology.com/evaporation-and-vapour-pressure/

Evaporation is the procedure of modifying fluid to fumes. This procedure

can be used to individual an assortment of fluids which have different
steaming factor temperature ranges. It is known that a fluid generates
fumes at any heat range. This is due to the molecular movement which
happens at any heat range above overall zero. The movement causes some
of the elements to ‘jump out’ of the fluid into the area above the fluid. The
higher the heat range, the more elements get into the fumes area.

The variety of elements in the fumes area cause a ‘Vapour Pressure’ above
the fluid and, again, the higher the heat range, the higher the fumes stress.
The fumes stress is termed as the ‘VP’ of the fluid at a particular heat range.
For any given fluid at a particular heat range, the variety of elements in the
fumes area is continuous, thereby providing a continuous VP

If we consider a genuine material like water at a continuous heat range, the

movement of the elements causes some to keep the fluid and get into the
fumes area developing a fumes stress. For example, the fumes stress of
water at 32 °F (0 °C) is 0.1 psia, well below environmental stress. At 100 °F

(37.8 °C) the VP is 1.0 psia and at 212 °F (100 °C) it is 14.7 psia (101 kPa). The
factor at which a fumes stress of a fluid is at its highest possible for the
particular heat range, the VP is known as the Soaked Vapour Pressure (SVP)
and, at this factor the fluid and its fumes are said to be at ‘Equilibrium’. (See
Figure: 1)

Figure : 1

At stability, the variety of elements coming into the fumes area is similar to
the variety coming back to the fluid. This situation continues to be
continuous as long as the heat range continues to be continuous.

Increase or loss of heat range will outcome in an improve or loss of the

variety of elements in the fumes stage and therefore an improve or loss of
the VP – but, when the heat range stabilises, the fluid and fumes will again
be at stability and display the VP for that heat range.

June 1st, 2012 in Vapor Liquid Equilibria

Distillation Tray Downcomer Calculations
http://seperationtechnology.com/downcomer-calculations/

Trillium Flow Technologies

Application-Specific Pumps
Complete Project Support Through Each

Stage Of The Life Of Your Operation.

Contact Us Now.

trilliumflow.com

OPEN
Downcomer Residence Time:
Sufficient residence time must
be provided in the downcomer to allow
adequate disengagement of vapour from the descending liquid, so that the
liquid is relatively vapour free by the time it enters the ray below.
Inadequate removal of vapour from the liquid may choke the downcomer.
 

Steps to calculation of Residence Time in Downcomer:

o Calculation of Average Downcomer Area: It is average of

inlet area (ADT)
and outlet area (ADB) of downcomer.
Average Area = ( ) / 2 DT DB A + A

o Calculation of Downcomer Volume

Downcomer Volume = AverageArea *
S /12 , where S: Tray spacing, inches

Calculation of Residence Time:

* DC Volume/ GPM;
Residence Time (tR), in sec. = 448.83

Compare the calculated residence time with recommended, if it is greater

than the downcomer volume is sufficient for disengagement of vapour for
the descending liquid, otherwise change the configuration.

Pressure Drop Calculation

Steps to determine Pressure
Drop:
i. Calculate Discharge Coefficient

Discharge coefficient (CV) is obtained from plot of Ah/Aa Vs discharge

coefficient at various ratio of t/dH
Where,

tt = tray thickness, in consistent units

dH = hole diameter, in consistent units

Ah = hole area, in consistent units

Aa = Active area, in consistent units

Calculation of Dry Pressure Drop:

The calculation of dry pressure drop

(hd) across the disperser unit is given
by a variation of the orifice equation:

Wet Pressure Drop (hl)

the aerated liquid is calculated from:
The pressure drop through
h Hydraulic gradient = 0 for sieve tray (Detail calculation Refer Kister
Section 6.3.3)

? = (aeration factor) is calculated from the graph below or equation given

below.

hW = Weir height, in
hOW = Crest over the
weir, in
QL = Liquid flow per unit weir
length(GPM/in)

iv. Total Pressure drop (ht)

Downcomer Backup:
Downcomer backup flooding occur when the backup of aerated liquid in
the downcomer exceeds the tray spacing

i. The factor causing Downcomer backup (hdc) are:

– Froth Height on the Tray

– Pressure Drop on the Tray

– Frictional Loss

Mathematically expressed as:

hdc = hc + ht + hda

Where,

iii. Calculate % Backup in the Downcomer

% Backup should be less than 80%. If it is high the changes in the design as
stated above can
be made.

June 14th, 2012 in Design Distillation System

Selection Criteria of Packings
http://seperationtechnology.com/selection-criteria-of-packings/

New Compressors

Available
Lab Gas Parts Ltd
A packed bed provides a mechanism for mass or heat transfer through
which the gas & liquid phases usually flow counter currently in the column.
The presence of tower packing elements provides a resistance to the flow of
these fluids that is greater than it would be in an empty column shell.
Packing tends to be used selectively for the following applications:

(3
a) Small tower diameters – When a tower diameter is less than 0.9m
feet), it is difficult to access the tower from inside to install and maintain
trays. Packing is normally a cheaper and more desirable alternative for
small diameter towers. However in special cases cartridge trays are used
for diameters less than 0.9m (3 feet)
b) Low pressure drop applications – Packing pressure drop is usually
much lower than trays. The open area of each tray is typically 5 % to 10 %
of the tower cross sectional area while the open area in a packed tower is
usually greater than 50 %.

range of packing materials is wider than that of
c) Corrosive systems – The
commonly available for trays. Ceramic and plastic packing is cheap and
effective for handling corrosive fluids.

packing is used to replace trays in
d) Existing tower revamps – Frequently,
order to increase the capacity of an existing tower.

classified as follows.
Based on the geometrical construction, packings are

a) Random Packings: These are the discrete pieces of packings of a specific

geometrical shape, which are simply dumped into the column. Random
packings are further classified according to their three generations of
evolutions. The first generation (1907 to 1950) produced two basic simple
packings, the Raschig ring & Berl saddles. These packings have all been
superseded by more modern packings & seldom used. The second
generation (late 1950’s to 1970’s) produced two popular packings- the Pall
ring, which evolved from the raschig ring & the Intalox saddle, which
evolved from the berl saddle. The third generation (the mid 1970’s until
present) has produced a multitude of popular geometries like IMTP, CMR,
Snowflake, Super Intalox saddle as etc.

Following are the points to be considered while selecting the proper type of
packing for particular service.
Specific surface area (m2/m3): This is the surface area per unit volume of
packing. This maximizes vapour liquid contact area & therefore efficiency.
Hence for random packings efficiency generally increases with decrease in
packing size. For structured packings efficiency increases as the space
between adjacent layers is decreased & for grid packings it increases as the
lattice opening are narrowed. Modern efficient third generation random
packings have specific surface area of 300 as compared to latest wire gauze
structured packings of 700 m2/m3.

Spreading of surface area uniformly: This improves the vapour liquid

contact & therefore efficiency. For e.g., a raschig ring & pall ring of similar
size have identical surface area per unit volume, but the pall ring has a
superior spread of surface area & is therefore much more efficient as
compared to raschig ring.

Random packing particles that interlock with
Uniformity of distribution:
other particles can lead to channelling & therefore lowers the efficiency.
Draining of liquid: Packing shall freely drain any liquid so that stagnant

liquid pockets are minimized. In case of latest type of saddles liquid pockets
are minimized which are there in older berl saddle type of packings.

Wetting rate: The minimum-wetting rate is the lower stability limit of

packings. It is the liquid load below, which the falling liquid film breaks up
& the liquid shortage causes dewetting of the packing surface. Dewetting of
packing surfaces at low liquid rates reduces efficiency & restricts the
turndown requirement.
Below are the values of minimum wetting rates for different packings
based on material of construction.

Voidage (Void fraction): Maximising the voidage minimizes the resistance

to vapour upflow & therefore enhances packing capacity by decrease in
pressure drop. Hence for random packings, capacity increases with particle
size. For structured packings, capacity increases with the space between
adjacent layers & for grids it increases as the lattice openings are widened.
Comparing efficiency & capacity of a packing, the ideal size of packing is a
compromise between maximising efficiency & capacity.
Material of construction: As already explained above in
6.2.e, packings
are available in different material of constructions ranging from metal
(carbon steel, stainless steel, copper, titanium), plastics (polypropylene,
PTFE, PVC), glass & ceramics (glazed & unglazed). Metal packings offers
higher capacity & efficiency, a wider range of geometries, higher turndown,
are unbreakable & have a high compression resistance. Ceramic packings
are specified for applications where their high chemical inertness and
resistance to high temperature gives them the edge (e.g. in sulphuric acid
absorption). Ceramic packings have lower capacity, breakable & are
unavailable in many of the popular geometries. Plastics (polypropylene)
packings are inexpensive & is most used when temperature do not exceed
122oC (250oF) – depending on its grade. Plastic packings tend to degrade in
oxidizing atmospheres & when solvents are used, & are avoided in such
application. A disadvantage of plastic packing is their poor wettability.
Friction: Minimising the friction favours as open shape that has good

aerodynamic characteristics, this in turn increases the capacity.

Strength: Packing shall have a maximum resistance to mechanical
deformation or breakage under the weight of bed.
Other than above points it should be of low cost &
shall have good
resistance against fouling & corrosion.

June 13th, 2012 in Distillation Technology

Designing Distillation – Section 1/4 – Basic concepts
http://seperationtechnology.com/designing-distillation/

Sepra Chem Corporation

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

This technical guideline is mainly intended for use of process engineers

involved in designing distillation column

PROCEDURE:

The steps involved in process design of fractionation column are:

1. Basic concept.

2. Laws ruling distillation

3. Azeotropes
4. Different models for distillation

5. Method for determining volatility, relative volatility.

6. Different types of distillation.

Basic Concept:

Distillation is a method of separating components of a solution which

depends upon the distribution of substances between gas & liquid phase
applied to case where all components are present in both phases.
Distillation is concerned with separation of solutions where all components
are appreciably volatile. In distillation new phase differs from the original
by its heat content. It is a method to separate components by virtue of their
differing boiling points i.e. volatility.

Laws ruling distillation:

DaltonsLaw:  Total pressure of system is equal to the summation of

partial pressures.

Where Pi is partial pressure.

For an ideal gas or vapor partial pressure is proportional to mole fraction of

constituent in vapor phase.

PA= yA × P

Where yA is mole fraction of constituent in vapor phase.

 Raults Law : For a an ideal gas mixture partial pressure is proportional
to product of pure component vapor pressure & mole fraction of
constituent in liquid phase.

PA= P°A× X A0

Where PºA is pure component vapor pressure.

Where xA is mole fraction of constituent in iquid phase.

This relationship holds true for higher value of xA.

For lower value of xA linear relationship exists between PA & xA   by a new
proportionality factor H’. This relationship is given by Henry’s law &
represented as

PA= H × X A

Where H is Henry’s constants

Azeotrope:

When the positive deviations from ideality are sufficiently large & the
vapor pressures of two components are not too far apart, the total pressure
curve at constant temperature may rise through maximum at some
concentration & such mixture is said to form an azeotrop or a constant
boiling mixture. Solutions like this sort cannot be completely separated by
ordinary distillation method. At azeotropic conditions y* = X & ? = 1.
Azeotropic composition as well as boiling point change with total operating
pressure.

Positive deviation from ideality

A mixture whose total pressure is greater than that computed for ideality is
said to show positive deviation from Rault’s law. In this case partial
pressures of each component are larger than ideal.

Two type of azeotrop are observed.

1)      Maximum Boiling azeotrope. : When the total pressure of system at
equilibrium is less than the ideal value the system is said to deviate
negatively from Rault’s law. When the difference in vapor pressure of the
component is not too great & in addition the negative deviations are large
the curve of total pressure against composition may pass through a
minimum. This situation may give rise to maximum boiling temperature
for azeotropic mixture. Maximum boiling azeotrop are less common than
minimum type.

2)      Minimum Boiling azeotrope: A mixture whose total pressure is greater

than that computed ideally to show positive deviation from Rault’s law.
When the positive deviation from ideality is sufficiently large & vapor
pressures of two components are not too far apart, the total pressure curve
at constant temperature may rise through a maximum at some
concentration.

Different models for distillation

Distillation models are of two types.

1) Constant pressure equilibrium.

2) Constant temperature equilibrium.

Consider a binary mixture A-B in which liquid component dissolves in all

proportions to form a homogenous solution which is not ideal & that no
minimum or maximum boiling point occurs. Component A shall be more
volatile than component B. Let x be the mole fraction of more volatile
substance A in liquid & y* be equilibrium mole fraction of A in vapor phase.
1)      For Constant pressure equilibrium

In diagram A upper curve provides temperature / vapor composition

relationship & lower curve represents temperature / liquid composition
relationship. Liquid & vapor mixture are at equilibrium throughout so that
horizontal tie line joins equilibrium mixture. Any point on lower curve
represents saturated vapors. Lower curve is called bubble point
temperature curve & upper curve is called dew point temperature curve.
Any point on upper curve represents saturated vapor.
2)      For Constant temperature equilibrium

This graph is similar to constant pressure equilibrium. A point above x vs P

curve represents entire liquid. If pressure is reduced at constant
temperature then first bubble shall occur on the curve x vs P. Hence it is
called bubble point curve. Complete vaporization shall occur along the
curve y* vs P, hence it is called as dew point curve. Any point below dew
point curve shall represent superheated vapors.

Method for deterring volatility & relative volatility

Volatility:  It is a ratio of partial pressure to the mole fraction in the liquid.

For a binary mixture A-B
Above equation can be used for calculation of composition of vapors for a
given x if ? is known. If ? is <= 1 then no separation shall occur.

For non ideal system activity coefficient ? is used.

Different methods of distillation

Broadly Distillation can be done as single stage or a multistage operation &
different methods used are as follows;

1) Simple Distillation

2) Flash Vaporization

3)  Continuous Rectification

First method is generally used for batch type & remaining two for
continuous type for binary & multicomponet mixture.

1)      Simple Distillation: Batch of liquid is charged to kettle .Kettle has

heating provision either by electricity or by steam. The charge is boiled
slowly & vapors are withdrawn to a condenser as rapidly as they form,
where they are liquefied. Condensate is collected in receiver. The first
portion of distillate shall be richest in more volatile substance & as
distillation proceeds the vaporized product becomes leaner. The distillate
can be collected in several separate batches called cuts. Consider a binary
system A-B of which A is more volatile than component B. Assume  that at
any time during the course there are L mol of liquid in the still of
composition x mole fraction of  A & that an amount dD mol of distillate is
vaporized, of mole fraction y* in equilibrium with the liquid . Material
balance equation can be written as

FX F = DyDAV + WXW

Operating equation can be written as follows &is called as  Rayleigh

equation.
 

Where

L = mol of liquid in still

F = mol of charge of composition xF

W = mol of residual liquid of composition xW

Y* = mol fraction in vapor phase

X   = mol fraction in liquid phase

2)      Flash Vaporization: In this operation liquid feed is heated in a
convectional tubular heat exchanger. The pressure is then reduced,   vapors
form at the expense of the liquid adiabatically & the mixture is introduced
in to a vapor liquid separating vessel. Feed is introduced tangentially into
the covered annular space. The liquid portion of the mixture is thrown by
centrifugal force to the outer wall& leaves from bottom f while the vapor
leaves from top. Material balance equation is given as

FZF = DyD + WXW

3)      Continuous Rectification: In this separation method part of condensed
liquid is returned as reflux & maximum enrichment of more volatile
component in vapor is obtained by successive partial vaporization &
condensation by multistage contact of vapor & liquid. This is achieved in a
single unit called fractionating column. Here feed is introduced more or
less centrally into the vertical cascade of stages. Section above feed tray is
called rectifying section.  More volatile vapors are contacted with liquid in
the form of reflux which results in the condensation of heavier fraction
present in it. The liquid return to top of tower is called reflux & material
permanently removed from top is called distillate which may be liquid.
Distillate is richer in more volatile component. Section below the feed plate
including feed plate is called stripping section. Liquid is stripped of volatile
component by vapor produced at bottom by partial vaporization of the
bottom liquid in the reboiler. The liquid removed from bottom is less
volatile component & is the residue. In the column liquid is always at its
bubble point & vapors are at dew point.
 
 

April 30th, 2012 in Design Distillation System

Distillation Design – Section 2/4 – Basics
http://seperationtechnology.com/distillation-design/

Over 25 Years Experience

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

This technical guideline is mainly intended to use of process engineers

involved in designing the distillation column the document covers design
consideration of distillation column, which includes:

– Distillation Design Procedure

– Operating Pressure

– Reflux ratio and Number of stages

– Determination of Number of theoretical stages.

– Feed location

This guideline covers the shortcut method to determine the above design
parameter

Specify Separation:

A material balance around the column is the first step in fractionation

calculations. In order to perform this balance, assumption of the product
stream compositions must be made. There are three ways to specifying
desired product from the fractionators:
 A percentage recovery of a component in the overhead or bottom
stream.

A composition of one component in either product.

A specific physical property, such as vapor pressure, for either product.

In a multicomponent mixture, there are typically two components, which

are “keys” to the separation.

Light Key (LK) Component: Defined as lightest component in the bottom

product in a significant amount.

Heavy Key (HK) Component: Defined as heaviest component in the

overhead product in a significant amount.

Normally, these two components are adjacent to each other in the volatility
list. For hand calculations, it is normally assumed for material balance
purpose that all components lighter than the light key are produced
overhead and all components heavier than the heavy product are produced
with the bottom product

Set Column Pressure:

Before any design calculations can be made on a fractionation problem, a

tower operating pressure must be determined. One of the primary
considerations for operating pressure is cooling medium available for
reflux condenser. The cooling media typically used are:

Air: The least expensive cooling method. Design limits the process to an
11°C approach to the ambient summer temperature.
Cooling Water: The satisfactory temperature approach is 5 to 10°C.

Mechanical Refrigeration: For process temperature below 35°C. This is

the most expensive cooling method from both a capital and operating
cost.

Example: If cooling water supply is at 32°C and return temperature of 40°C

using above guideline the condensing temperature is 45°C. For total
condenser with a distillate composition of C1, C2, C3 as 5/5/90%, the vapor
pressure of the distillate @ 45°C is 23.3 kg/cm2a.   Set the pressure of the
reflux drum equal to or slightly above the vapor pressure of the distillate at
45°C.

Generally, it is desirable to operate at as low a pressure as possible to

maximize the relative volatility between the key components of the
separation.  For a quick estimate of a gas oil/resid operation a pressure of
50 mm Hg at the overhead line and a flash zone temp of 370°C may be used.
However, if reducing the pressure requires a change to a more expensive
cooling method, this is the usually not a desirable choice.

Whereas, in case of lighter hydrocarbon separation, say C2 from C3, use of a

suitable refrigerant is required when the distillate liquid temperature is
lower than the cooling water temperature. Designing such a column at the
higher pressure will make:

The refrigerant system smaller, but

Reboiler duty and the column diameter will be larger.

In deciding the operating pressure of a column, a number of factors have to

be considered in relation to the total system.  A table is summarized below
to indicate the advantages and disadvantages of operating a column at
higher pressure.

 Advantages

It increases the boiling points of the distillate and enables a cheaper

cooling medium to be used.  If the cooling medium remains unchanged,
it reduces the condenser area requirements.
It reduces energy costs in vacuum operation.

It increases vapor density and vapor handling capacity. Leads to smaller

column diameter for pressure upto 3.5 to 10.5 Kg/cm2.

It increases boiling points of liquids in the column for distillation of

liquefied gases and allows the use of cheaper construction materials.
 It reduces the size of pipes and valves for the vapor line.
Below 1 atm, reduces air leakage into the system.

Disadvantages

It decreases relative volatility and makes separation more difficult.  It

will therefore increase reflux and stage requirements, and requires
higher reboiler and condenser duties.
It increases bottom product temperature and increases the risk of
chemical degradation, polymerization and fouling.

Above 7 kg/cm2g, column shell thickness will probably increase,

whereas below 7 kg/cm2g, the effect is not as significant.

Increases reboiler temperature, needs a more expensive heating

medium or increase in reboiler area requirement.
It increases hazard potential if process materials are inflammable or
toxic.

For mature technologies the operating pressures of the column have been
standardized except for some variation due to site condition, reference can
be taken from similar plant/equipment.

Determine minimum reflux and minimum number of stage 

Relation between Reflux Ratio and Number of stages:

The design of a fractionation column is a capital cost versus energy cost

trade-off problem. The primary parameters are the number of stages and
the reflux ratio. For most calculation, reflux ratio is defined as the ratio of
the molar rate of reflux liquid divided by the molar rate of net overhead
product. The desired separation can be achieved between the limits of
minimum reflux and minimum stages. The relationship between reflux
ratio and number of stages for a given separation is shown on figure 1
below.  At minimum reflux an infinite number of stages are required.  At
total reflux a minimum number of stages are required
 

Determination of Minimum Number of Stages 

Fenske’s equation can be used to calculate the minimum number of stages

 

Determination of Minimum Reflux Ratio

The Underwood method can be used for calculating minimum reflux ratio. 
The Underwood method assumes:

Constant molar overflow, and

Constant relative volatility at the average column temperature.

The first step is to evaluate? (Which must lie between the relative
volatilities of the keys) by trail and error from
Once s is determined, the minimum reflux ratio is

 
 

Where,

Lo = Liquid Reflux Rate, kgmole/hr.

D = Distillate Product Rate, kgmole/hr

T     =     Correlating Parameter

a   = Relative Volatility

i    = any component

X   = Liquid Rate, kgmole/h

Set reflux ratio and calculate number of theoretical stages

It is customary to always add a certain safety margin above the minimum

reflux level to avoid potential errors from VLE estimation or from
misdistribution at the desired reflux ratio.  The rule of thumb for R/Rmin
which is regarded satisfactory as a general guideline is listed below

There are several considerations in deciding the reflux ratio; the following
comments indicate the effect of a change in the reflux ratio in relation to
the number of stages.

The higher the reflux rate, the higher the cost of condensing and
reboiling. These make up the bulk of the column operating costs.
 As reflux ratio increases, the number of stages decreases, but column
diameter increases.

Close to minimum reflux rate, small increases in reflux ratio will reduce
the number of stages considerably and therefore column height, but
produce only a small increase in column diameter.

Further increases in reflux will have a smaller effect in the height

reduction but greater increase in column diameter.

The closest possible approach of design to minimum reflux requires

accurate methods such as vigorous stage wise calculations, developed by
multiple cases on computers.

Determination of Number of theoretical stages:

The number of theoretical stages required for a given separation at a reflux

ratio between minimum and total reflux can be determined from Erbar and
Maddox Correlation. This correlation relates the ratio of minimum stages to
theoretical stages to the minimum reflux ratio and the operating reflux
ratio as shown below:

We have calculated

Minimum Reflux Ratio = Rm = (Lo/V) m

Minimum number of stages = Sm

R = (Lo/V) = Reflux Ratio.

Knowing Sm/S from the graph below we can calculate S = Number of stage
Adjust actual reflux for feed vaporization if necessary

The Erbar and Maddox correlation is based on bubble point feed.  For a
feed stream between the bubble point and dew point the following
relationship adjusts the vapor rate from the top tray for non bubble point
feed.
 

V = Vapor rate, kg mol/hr

D = Overhead product rate, kg mol/hr

F = Feed rate, kg mol/hr

HVF = Vaporized feed stream enthalpy, Kcal/Kg mol

HBP = Bubble point feed stream enthalpy, Kcal/Kg mol

QC = Condenser duty Kcal/hr

LO = Liquid condensed in a partial condenser, Kg mol/h

Determination of feed location

Once a base case tower has been simulated, optimization studies on trays
vs. reflux, feed tray location, etc. can be carried out.  One criterion often
used for optimum location of the feed tray is to try and match the feed
composition as closely as possible to a corresponding tray composition. The
approximate feed location can be determined by the ratio of the total
number of theoretical stages above and below the feed plate from following
equation:

Fenske’s Equation

Kirkbridge’s Equation
Akashahetal’s Correction

Fenske’s Equation

 
 

(10)(eq. cont.)

 
Column Internals Explained : Part – 1
http://seperationtechnology.com/column-internals-explained-part-1/

Over 25 Years Experience

Horizontal & vertical coalescing filters, dual closure

coalescing filters, dust filters.

Sepra-Chem Corporation Open

In the recent years, efforts have been made to design structured & random
packings for mass transfer columns more & more effectively. These
developments are done to achieve greater throughputs; lower pressure
drops & better mass transfer efficiency. Column internals are also playing
important role for achieving the above targets & hence needs to be
designed carefully from both process & mechanical point of view. After the
packed column hydraulics, column internals needs to be designed. Packing
internals include support plates, distributors, redistributors, and bed-
limiters, feed pipes, collector trays & mist eliminators.

a) Support Plate: The primary function of the packing support plate is to

serve as a physical support for the tower packing plus the weight of the
liquid holdup. The packing support plate must pass both the downwardly
flowing liquid phase as well as the upwardly flowing gas phase.

Support Plate

For random packings, gas injection type of support plate is used. Gas
injection type of support plate provides the separate passage for liquid &
gas. Support grid (made up of bars) is used to support the structured
packing. Support plates have a very high open area for gas & liquid passage
& do not add any significant pressure drop.

b) Bed Limiter: Bed limiters are commonly used with metal or plastic
packings. The primary function of these devices is to prevent expansion of
the packed bed as well as to maintain the bed top surface level. In towers,
the packed bed will not fluidize over the entire surface, vapour surges
fluidize random spots on the top of the bed, so that after return to normal
operation the bed top surface is quite irregular. Thus the liquid distribution
can be affected adversely by such an occurrence.

Bed Limiter

Hold-down plates are weighted plates used with ceramic or carbon tower
packings. With these packings, it is especially important to prevent
fluidisation of the packed bed top surface. The hold down plate must rest
freely on the top of the packed bed because beds of ceramic & carbon
packings tend to settle during operation. These plates usually act by their
own weight to prevent bed expansion. Weight (kg) of the hold down plates
can be calculated by the pressure drop across the packed bed (kg/m2)
multiplied by the column cross sectional area (m2). They weigh ~ 95-120 kg
per sq meter (20 lb to 30 lb per sq ft).

c) Collector trays: Whenever there is partial or total draw off in between

two packed beds, collector trays are used. However it has been noted that
some packing vendors are always giving collector trays in between two
structured packing beds considering the high turbulence velocity of liquid
in structured packing.
Collector Tray
Since this type
of tray frequently maintains a fairly high level of liquid (and
consequently a tremendous weight), special consideration should be given
to its design. One consideration is the placement of the draw nozzle. A flat
chimney tray can be used with the nozzle located at tray floor level, or a
portion of the tray floor can be lowered to form a sump and the nozzle
located at the sump floor. Both designs require the same liquid head to
force the design flow rate through the nozzles. Therefore, locating the
nozzle in a sump lowers the liquid level on the chimney tray by an amount
equal to the head requirements. This reduces the weight, which the tray
must support, but has little effect on residence time since that portion of the
liquid depth corresponding the head requirements should not be
considered as residence time in most cases. A chimney height of 6 – 12″ is
normally adequate for low liquid flow. 12 – 18″ is usual for high liquid flow
if the draw nozzle is

located in an inlet sump. All chimneys should have hats located a sufficient
space above the chimney to give a peripheral area of 1.25 times the
chimney area. If a leak-free design is required, the inlet sump should be
seal-welded and gasketing used on the chimney tray floor. A centre sump is
preferred over a side sump, as some flow restriction can occur at very high
liquid rates due to the shorter weir lengths of side sumps.

February 18th, 2012 in Design Distillation System

Column Internals Explained – Part 2
http://seperationtechnology.com/column-internals-explained-part-2/

Over 25 Years Experience

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

d) Liquid distributor: From a process point of view liquid distributors are

the most important column internal devices. A distributor is required to
uniformly distribute liquid at the top of a packed tower. Liquid distributor
must provide sufficient gas passage area to avoid a high pressure drop or
liquid entrainment.

An ideal distributor possesses the following attributes.

Uniform liquid distribution

Resistance to fouling

Proper turndown capability

Low gas pressure drop

Minimum height

capability
Higher cross-mixing
Distribution quality: Quantifying the uniformity of liquid distribution is
accomplished by calculating the distribution quality (DQ) of a distributor
based on the Moore & Rukovena model. It relates the liquid load across the
column area at the top of packed beds marking circles proportional to the
liquid flow through a particular orifice & then considering the irrigated,
overlapping, & un-irrigated areas of the circles.

An ideal liquid distributor should have a DQ of 100%, but practical

considerations restrict the DQ to about 95% maximum. As per Moore &
Rukovena model, a low DQ indicates a high degree of liquid misdistribution
& some portions of the column cross sectional area may be receiving
substantially different volumes of liquid when compared to other portions
of the cross sectional area. As the DQ decreases the number of stages that
can be realized from a packed bed decreases, thus decreasing the
separation efficiency.
Following guidelines
are used for deciding the no. of points per m^2 for
mentioned latest type of packings.

Classification of liquid distributors: Liquid distributors are usually

classified into pressure & gravity distributors.

Pressure Distributors: This type of distributors
provides more open area for
vapour flow & tend to be less expensive, lighter, less robust & require
smaller lead up piping than gravity type. Disadvantages are high operating
cost, susceptibility to plugging & corrosion & a relatively inferior quality of
distribution. Pipe arm & Spray type liquid distributors are the examples of
pressure distributors.

Gravity Distributors: These are further classified as weir type & the orifice
type. Both of these types can handle large liquid flow rates. The weir type is
generally one of the least troublesome distributors, handle large liquid load
& has excellent turndown capability. However it can usually provide only a
limited no. of drip points (minimum 30 no. of weirs per m^2), extremely
sensitive to levelness & liquid surface agitation.

Orifice distributors are usually of the pan/deck type or of the tunnel

(trough) type. An orifice pan distributor consists of a pan equipped with the
circular or rectangular risers for vapour flow & perforations in the pan for
liquid flow. Orifice trough type distributors consist of parallel troughs with
perforations in the trough floors for liquid. Vapour rises in the space
between the troughs. The troughs are often interconnected by cross
channels (called as parting box) that equalize liquid levels in different
troughs. These trough type liquid distributors provide more open area for
vapour flow & are easier to support.

& plugging, but it can be designed
Orifice type may suffer from corrosion
with a large no. of drip points to provide superior liquid distribution.
Orifices are sized to maintain a minimum liquid head at desired turndown
conditions & to avoid distributor overflow during turn up condition. Very
small orifice diameters (<4mm) are avoided to prevent fouling.

Selection of a liquid distributor: Specific liquid distributor for particular

service is chosen by considering the tower diameter, flow rate, turndown
ratio, available height (in case of revamp) & fouling resistance.

Please refer the packing vendor catalogue for more details on liquid
distributor models/types.

Following types of liquid distributors to be used in applications that are less

demanding on packing performance, such as heat transfer applications,

absorption & stripping services.

Following are the some important design points of liquid distributor to be

checked before finalising its design.

Maximum & Minimum head: Maximum liquid head is the actual liquid
head on the deck of the distributor at turn up condition. The overall height
(riser/trough) of the liquid distributor shall be higher than this head by
following formula.

Riser (Trough) height = {(Maximum head) 1.10 + 25} mm

It is always required to estimate the minimum head on the distributor at

turndown condition. At low liquid head there are chances of liquid
vortexing & entrainment due to insufficient resistance to orifices. Also if
minimum heads are fallen short of, considerable misdistribution can occur.
Hence minimum liquid head on the distributor shall be around 35mm. It is
always recommended to use punched orifices in the direction of liquid flow
instead of drilled orifices. Pressure Drop & open area requirement for
vapour: Liquid distributor must provide sufficient gas passage area as per
gas load to avoid a high-pressure drop or liquid entrainment. In case of
very low open area required for gas, a minimum of 12% open area is used.
Pressure drop of liquid distributor shall not exceed a value of 20mm WC.
However it will depend upon gas load, liquid load & a particular service.
% Flow variation: For low liquid loads & small column diameters, flow

variation will increase due to low liquid head. Flow variation is a strong
function of mechanical points such as plate levelness, distributor deck
flatness as etc. Excessive flow variation will cause liquid misdistribution on
below packed bed, entrainment & hence needs to be minimised. Maximum
of 12% flow variation is allowed for liquid distributor or redistributor.

March 20th, 2012 in Design Distillation System

Column Internals Explained – Part 3
http://seperationtechnology.com/column-internals-explained-part-3/

Sepra Chem Corporation

Horizontal & vertical coalescing filters, dual

closure coalescing filters, dust filters.

Sepra-Chem Corporation

Liquid redistributors: One of the main functions of a liquid redistributor

is to remix the liquid phase so as to bring the entire liquid flow onto the
next lower bed at a more uniform composition.
Deck type Liquid Redistributor
If liquid tight pans or boxes are used for a redistributor, gas riser covers &
wall wipers must be used. Whenever the liquid falls directly onto a deck
type redistributor, the gas risers must be provided with covers to prevent
liquid from raining into this area of high vapour velocity. All covers should
have located a sufficient space above the risers to give a peripheral area of
1.25 times the
riser area.

Process demand various feed to be introduced into the column
Feed pipes:
at various locations. The feed being introduced could be liquid only, liquid
& vapour above a packed & vapour only below a packed bed.

Liquid Feed Pipe

devices, liquid is fed into the distributor via feed pipe &
For liquid only feed
feed pipe design depends upon distributor type, liquid flow rate, its
operating range as etc. Each feed pipe meters flow to one or more
appropriate feed areas, matching the hydraulic requirements of the
distributor. Excessive turbulence, hydraulic jump & horizontal flow velocity
in the liquid distributor are eliminated in a feed pipe design. A submerged
feed is recommended in all cases in large towers with low or high flow
rates. Submerging the feed maintains uniform liquid temperature &
reduces hydraulic jump of liquid in the distributor. Liquid feed pipes are
designed with a velocity of 1m/s to max 3.5m/s.

For liquid & vapour feed devices above a distributor, separating the two
phases is of primary response. The primary design factors are the feed flow
rate, desired turndown, column height needed for flashing vapour
distribution & mixing of the inlet liquid with overhead liquid. Flashing feed
devices such as flash feed gallery, flash feed chamber or flashing feed pipe
are used based on the composition of vapour in liquid.

Flashing feed pipe

Vapour only feed devices are required for reboiler returns or to introduce
vapour feed or to introduce vapour or gaseous feed. If the column offers
adequate pressure drop (> 4-5 mm of WC per m of packed bed height), the
packings themselves tend to mix the vapours. In the event of very low
pressure drop across the column (i.e. < 4-5); vapour channelling can
become a serious problem. The kinetic energy of the vapour (F factor) & its
composition at the point of introduction are the two main factors
considered in designing the vapour entry device. Vapour or gas feed pipes
are designed with a velocity of max 20m/s.

ii) Vapour composition: When the vapour is introduced between packed

beds, consider the degree of mixing of the inlet vapour with rising vapour.
If a gross mismatch of composition &/or temperature exists, mixing of the
two vapour optimizes packing performance above.

g) Mist Eliminator: In every process involving contact between liquid and

flowing gas, tiny mist droplets are carried away with the gas. This
phenomenon is called entrainment. Mist eliminator provides a large
surface area in a small volume to collect liquid without substantially
impeding gas flow. It work on the principles of inertia whereby small
droplets of liquid entrained in

drops & fall downward due to gravity.
vapours are coalesced into large
March 25th, 2012 in Design Distillation System
FILTERS & STRAINERS
http://seperationtechnology.com/filters-strainers/

Trillium Flow Technologies

Application-Specific Pumps
Complete Project Support Through Each

Stage Of The Life Of Your Operation.

Contact Us Now.

trilliumflow.com

OPEN

During the circulation of liquids through procedure sewerlines and devices,

strong contaminants of dirt, corrosion, range.. etc, may be grabbed by the
liquids. These shades can cause damage to devices by break down and
obstruction. To be able to reduce this problem, Filtering system or Strainers
are set up in pipes, upstream of devices in which strong contaminants are
unwanted.

As an example, strainers are set up upstream of some steam-traps which, if

strong contaminants were to get into the snare, its function would be
impacted negatively.

Filters are designed according to the responsibility they are to execute – a

good example is the air narrow in the consumption to a car motor. It is
created of very fine content to be able to eliminate very small contaminants
of dirt which would cause problems in the carburettor where the air is
blended with the petrol. The petrol is also strained before combining.

To secure the motor bearings, golf club shafts, equipment and aide …etc,
the car lubricant oil is also strained. In all cases where filters or strainers
are used, they must be washed or the unclean components modified for
fresh ones as soon as the filtration procedure becomes inadequate and fluid
circulation reduces.
1. FILTERS (Figure: 24)

Filters used in procedure functions to eliminate strong contaminants, are

designed of a number of components created up of cable supports or
pierced steel pipes wrapped by perfectly weaved fabric or steel cables
(screens).

Other kinds contain fibre-glass components which adsorb the contaminants

of dirt, range and corrosion. Still further kinds involve ‘basket’ kind
components which can easily be eliminated, washed and changed. The
narrow components are placed in a round boat which is linked into the
pipe-work holding the fluid.
In ongoing procedures, two narrow models may be set up, where one is in
use while the other is on ‘standby’. A total narrow avoid is also usually fixed
for use should both models become fouled due to unwanted shades in the
program not enabling washing or factor alternative to take place. However,
in such a situation, filters must be re-commissioned as soon as possible.

The program will have a means of calculating the stress fall across the
narrow and, when this D.P. gets to a certain point, an alert change may be
triggered. The filters must then be modified over, the unclean narrow is
then separated, de-pressured, cleared and started out up to eliminate the
unclean components. New components are then set up and it ‘boxed up’.

The fresh narrow is then filled again and/or re-pressurized by using the
little collections and valves offered. While reloading, air is venting to make
sure that the narrow is liquid/gas full. It is then placed on ‘standby’. Where
a single narrow device is set up, again, the avoid is used while washing is
performed.

2. STRAINERS
Where little pipes like petrol resources, air to equipment and vapor /
condensate techniques are worried, strainers are set up to make sure the
sleek function of equipment and vapor barriers by eliminating strong
contaminants.

The main kinds of strainer used in such techniques are :

THE ‘Y’ STRAINER.

THE ‘T’ STRAINER

THE ‘CONE’ STRAINER.

The following images show the application of strainers in procedure

functions.
Over 25 Years Experience
Horizontal & vertical coalescing filters, dual

closure coalescing filters, dust filters.

Sepra-Chem Corporation
For pressure drop estimation graphs are referred. This graph gives pressure
drop based on specific gravity of water 1.0 and viscosity of water 1.0 cP.
Screen openings chart indicates the % open area of perforated plate and
meshed line screens. For fluids of high viscosity and finer weave meshes
viscosity and density correction factor chart will be referred.

Sample calculation:

across 100 mesh lined Basket Type strainer
Estimate pressure drop
provided in a 6″ line with following data.
Flow rate = 700GPM

Viscosity = 100cP

Specific Gravity =
1
The Pressure Drop
Chart for Basket strainers indicates a drop of 0.9psi.

The Screen openings chart indicates the % open area of 100 mesh is 30%.
From chart one correction factor to be 1.2.

Total Pressure drop (P1) = 0.9 X 1.2

= 1.08 psi

by the specific gravity of the fluid actually flowing through the
Multiply P1
strainer to get P2.

=1
Since specific gravity
Pressure drop (P2) = P1 X
Sp. Gravity
= 1.08 X 1
=1.08psi

2 multiply P2 by the appropriate component factor (CF) to get
Using chart
P3.

chart 2, (P3) = P2 X (CF)
Using
= 1.08 X 0.35

= 0.38psi

Let, P4 =
P2 – P3 = 1.08 – 0.38
= 0.7psi

Multiply
P3 by the appropriate body loss factor (BF) in chart 3 to get P5
Using chart 3, P5 = P3 X (BF)

= 0.38 x 1.6

= 0.61psi

Multiply P4 by the appropriate Screen Loss factor (PF OR MF) in chart 3 to

get P6
Using
chart 3, P6 = P4 X (PF or MF)
= 0.7 X 6.5

= 4.55psi

Total pressure drop (P7) = P5 + P6

Total pressure drop (P7) = 0.61 + 4.55

= 5.16psi

Annexures:
 
February 14th, 2012 in Other Unit Operations
General steps for packed column hydraulics
http://seperationtechnology.com/packed-column-hydraulics/

Trillium Flow Technologies

Outstanding Flow Control
Our Products Are Backed By Expertise &

Support. Contact Us Today To Learn More.

trilliumflow.com

OPEN

Before doing the packed column hydraulics separately, it is assumed that

the necessary hydraulic calculations have already performed in column
sizing mode along with process design of distillation column, absorber or
stripper on simulation software such as Hysys, Chemcad, Pro-II etc.
However generally packings & internals vendors perform hydraulic design
of packed column & include the hydraulic guarantee in their scope.
Following guidelines are useful in case we have to size the column or check
the packed column hydraulics.

Following are the general steps for packed column hydraulics:

The results of process calculations for the column from simulation package
are made available to the concerned engineer in the form of stage wise
loading summaries & its properties, which will be generated in column
environment of simulation software such as Hysys, Chemcad, Pro-II as etc.
These loadings are used to generate the PDS for packed column.

Packed bed height & total no. of beds needs to be finalised based
on the
HETP (with desired safety margin for particular service) of selected
packings. Theoretical no. of stages getting after doing process design of a
particular column on simulation software needs to be converted into actual
no. of stages or actual packed bed height) based on HETP concept.
Particular type of packing to be selected based on the guidelines as given
above or from past experience on similar column.
For e.g. for 5 no. of theoretical stages for IMTP#15 packings, total bed height
will be HETP times the theoretical no. of stages. As approximate HETP value
of IMTP # 15 packing is 350mm. Total packed bed height will be calculated
as below.

Total packed bed height = HETP X theoretical no. of stages

Hence total packed bed height for above case will be 1750mm.
HETP & other physical properties such as surface area, packing
factor, void
fraction as etc of different packings are to be taken from packing vendor or
from open literature as given in reference documents.

be infinite in
There is no conclusive evidence that packed bed may not
depth, from a process point of view. There are however mechanical
limitations such as crushing or deformation exists for deep beds. Maximum
packed bed height for different packings is the ratio of maximum allowable
load (kg/m2) to its density plus liquid hold up in kg/m3. Generally maximum
bed height of 8m is recommended for random & structured packings.

c) Column diameter to be calculated based on the allowable pressure drop

(defined at the time of process design) & %flooding across the column.
Absorbers, strippers & direct contact coolers are generally designed with a
pressure drop of 33.3 mm of WC per m (0.4 inch of WC per feet) of packed
bed height.

d) After fixing the column configuration (diameter & packed bed height),
internals such as liquid

collector tray, feed pipe, as etc needs to be
distributor, redistributor,
designed.

March 13th, 2012 in Design Distillation System

Packing Hydraulics Calculations
http://seperationtechnology.com/packing-hydraulics/

Trillium Flow Technologies

Application-Specific Pumps
Complete Project Support Through Each

Stage Of The Life Of Your Operation.

Contact Us Now.

trilliumflow.com

OPEN

There are several software programs for packed bed hydraulic calculations
released by vendors, some of them have been installed on their website
also. These programs, vendor brochures or column-sizing mode in
simulation software’s can be used for process studies and preliminary
designs. For final designs, the Process Engineer should obtain the hydraulic
calculations from packing vendors.

Flooding %: Percentage of flood is normally used to indicate the degree of

approach to the vapour rate at flooding point or the maximum capacity of a
packed bed. When a tower is operating at a higher percentage of flood, it
has a better efficiency. However, the liquid entrainment may become
excessive if the percentage of flood is too high. Flooding curve between
flow parameter & vapour momentum force is used to calculate the %
flooding across the packed column. Flow parameter is the ratio of liquid to
vapour kinetic energy.

Procedure for calculation of % flooding across the packed column:

i) Calculate the flow parameter (abscissa) value by following formula.

ii) Ordinate value to be taken from for upper flooding curve from above
chart & calculate the flooding vapour mass velocity by following formula.

each type of packings. Lower packing factor values means higher the
capacity & lower the efficiency of a particular type of packing. Kindly refer
table 6.3, (page no. 196) of Mass Transfer Operations of Treybal for packing
factor.

iii) % of flooding can be calculated by the ratio of normal vapour mass

velocity (G’) to flooding vapour mass velocity (G) for a given column
diameter.

It is recommended that new towers be designed at a maximum of 80 % of

flood. A higher percentage flood of around 85% may be used in tower
revamps.

March 13th, 2012 in Design Distillation System

Distillation Columns in Fouling Service – Design
Guidelines
http://seperationtechnology.com/design-guidelines-for-distillation-columns-in-fouling-service/

Sepra Chem Corporation

Wet filters, 3N1 coalescing filters, salt water

removal, & more. Call now.

Sepra-Chem Corporation

Designing distillation columns in fouling service requires first an

understanding of the fouling mechanism, the process in which the fouling
occurs, and behavior of the process when the fouling is present. An
understanding of these items needs to be developed in advance of
designing mass transfer equipment for fouling service.

The challenges of fouling columns can result in;

1. Increase energy consumption due to heat transfer and efficiency issues.

2. Reduced column capacity, which may lead to production loses.

3. Increased down time for cleaning and disposing of fouling wastes

4. Potential need for the use of chemical additive

General Guidelines

Fouling service in a distillation column is a broad term that encompasses

many fouling phenomena. The fouling phenomena can be localized in
either the vapor phase or liquid phase. It the fouling is localized in the
liquid phase an inhibitor can be utilized.

Fouling phenomena can include;

 1. Vaporization of volatile components
2. Polymerization
3. Condensation

4. Sedimentation / Precipitation / Crystallization

5. Foaming
6. Chemical Reaction

7. Corrosion

Complex fouling problems are a result of a combination of two or more of

these phenomena occurring at the same time, such as in an ethylene caustic
tower. They are nearly always mutually reinforcing.

Factors that assist fouling phenomena can include;

1. Residence time
2. Stagnate zones

3. Sharp transitions
4. Emulsion issues

Consideration must also include the fact that fouling gets worse over time.
Fouling is not static.

Fouling Mechanisms

Vaporization of Volatile Component

In many applications fouling can result if the volatile components are

allowed to vaporize. This fouling is found in refinery vacuum and de-
asphaltene oil towers and ethylene quench oil towers. If the process is
placed on recirculation without the addition of fresh feed with volatile
material, the column can foul as the volatile components are removed. In
one example an ethylene quench oil tower was placed on circulation and
viscosity of the bottoms product increased beyond its pour point. The unit
was shut down and required several days of off-line cleaning.
Polymerizatio

Polymerization is the linking of double bonds to form long chain molecules.

Examples of polymerization include, polyethylene, polypropylene,
polystyrene, poly butadienes and polyMMA. These products are
undesirable in the monomer distillation column. These products can lead to
reduced capacity and unit outages.

Condensation

Condensation is a reaction where two or more small molecules combine to

from large, stable structure molecules. Extreme condensation is the
formation of coke at high temperature and long residence times. Coke
forms when thermal cracking removes hydrogen and light material from
the fluid. A layer of coke is left behind. Thermal cracking can occur in both
the gas phase and the liquid phase. However, the most common situation is
liquid-phase cracking when low liquid rates extend residence time in high-
temperature operations.

Examples of condensation can be found in ethylene furnace transfer lines

to the quench oil tower and in refinery vacuum towers

Sedimentation / Precipitation / Crystallization

Sedimentation is the accumulation of solids that are deposited in low

velocity areas in process equipment. The equipment can include heat
exchangers, tower distributors, distillation trays, random packing, and
structured packing. If the trays are fouled below the feed point the cause
might be sedimentation.

Many processes contain suspended solids, which can settle out on the mass-
transfer surface. Products of sedimentation can include salts, metal oxides,
catalyst fines, and coke fines. Precipitation and crystallization of dissolved
salts can occur when process conditions become super saturated, especially
at mass transfer surfaces. Ammonia salt deposition resulting from both
water vaporization and direct solid deposition from the gas phase is a
common refining problem.
Sometimes the deposits formed do not adhere strongly to the surface and
are self limiting, the thicker a deposit becomes, the more likely is to be
removed by the fluid flow and thus attain some asymptotic average value
over time. Sedimentation fouling is strongly affected by velocity and less so
by temperature, however a deposit can bake on a surface and become very
difficult to remove.

Precipitation and supersaturation are fouling issues. Certain salts such as

calcium sulfate are less soluble in warm water than cold. If such a stream
encounters a wall at a temperature above that corresponding to saturation
for the dissolved salt, the salt will crystallize on the surface.

Crystallization will begin at specially active points (nucleation sites) such as

scratches and pits, often after a considerable induction period and then
spread to cover the entire surface. The build up will continue as long as the
surface in contact with the fluid has a temperature above saturation. The
scale is strong, adherent and can require vigorous mechanical or chemical
treatment for cleaning.

The following processes may cause super saturation:

1. Evaporation of solvent
2. Cooling below the solubility limit for solution with normal solubility,
solubility increases with temperature.
3. Heat above the solubility of solutions with inverse solubility such as;
CaCO3, CaSO4, CA3(PO4)2, CaSiO3, Mg(OH)2, MgSiO3, Na2SO4, Li2S04
and Li2Co3 in water.
4. Mixing of streams with different composition.
5. Variation of pH, which affects the solubility of CO2 in water.

6. Ammonia salt deposition in refining.

Additionally, solidification fouling can occur due to cooling below the

solidification temperature of a dissolved component, such as solidification
of wax from crude oil.
Knowing the properties of the fluid and suspended solid, flow conditions
and mixing effects one can develop guidelines to predict sedimentation of
solids from liquids. Petroleum fluid is a mixture of various compounds.
Solubility balances among all such components make a petroleum fluid
stable. Changes in that balance can cause precipitation from a petroleum
fluid and result in fouling (asphaltene precipitation)

Foaming System

Some systems have a propensity for fouling in the form of foaming. In

many systems an operating froth can be observed on the liquid phase. In
distillation systems, the decrease in surface tension as equilibrium
temperature rises promotes foaming. High gas velocities allow liquid to be
entrained into the vapor phase. In non-foaming systems liquid
disengagement from the vapor stream is relatively easy. Foaming makes
liquid disengagement from vapor flow difficult. When foaming becomes
severe it can lead to a reduction in capacity and loss of efficiency. In order
to counter this effect distillation equipment that operate in foaming systems
are run a lower vapor and liquid rates to reduce the amount of froth
generated. Sometimes anti-foam additives are added to distillation columns
to decrease the amount of foam generated in the column. There are
drawbacks to using anti-foam chemicals. Sometimes they may contaminate
the end product and produce product that is does not meet production
specifications. (7).

Chemical Reaction

In distillation desired or undesired chemical reactions can occur. In an

ethylene caustic tower there are competing chemical reactions. The caustic
absorbs the CO2 (desired) by electrostatic interactions and Van der Wall
attraction. The acetaldehydes can be polymerized by the caustic (undesired)
to form first a yellow oil polymer, and then a red oil polymer by aldol
condensation reactions, which can then lead to emulsions.

Corrosion

When corrosion occurs the increase surface roughness may promote

fouling from the other fouling mechanisms. A comprehensive theoretical
discussion of the different fouling mechanism is given in [4]

May 2nd, 2012 in Design Distillation System

Measures to Mitigate Fouling Phenomena
http://seperationtechnology.com/measures-to-mitigate-fouling-phenomena/

Workflow management

for AM
Authentise Inc.

Process Review

The first step in mitigation of fouling phenomena is to thoroughly review

the process for fouling potentials. Investigations of previous applications
need to be reviewed to develop the best solution to the specific process.

Equipment Selection

The most suitable mass transfer equipment for fouling service may also be
the least efficient for mass transfer. Grid packing and shed decks can
handle nearly every fouling service, but have low efficiencies when
compared to sieve trays, random, and structured packings.

Packing

For packed towers the key fouling factors revolve around liquid
distribution and packing residence time. The longer the residence times the
less suitable. Low-pressure drop, smooth surface, low residence time
packings perform best in fouling service. The order of preference is:

1. Grid
2. Structured packing
3. Random packing
Packing Distributor Concern

In fouling service distributors are areas where residence time is increased

and fouling phenomena can occur. In high-fouling services trough v-notch
or other type of trough distributors are recommended over pan type
distributors. A review of feed piping is also warranted.

Tray

The industry prefers trays in fouling service because of the long history of
success trays have had in fouling service applications. The first continuous
distillation column with bubble cap trays was developed in 1813 and
structured packing was developed in 1964. The database and application
know how is much larger with trays. The best trays to use in fouling
services are sieve trays and dual flow trays. Moveable and fixed valve trays
are less resistant to fouling because the valves are areas where fouling can
seed and propagate. Distillation is part art, part science. As in any science,
as you explore the limits you will find the

Example of Trays in Ethanol Mash Service.

Tray Downcomer Concerns

In fouling services downcomer design requires attention to detail. A major

problem with downcomers is the creation of dead spots: in areas near the
tower wall, in areas opposite the downcomer outlet, areas near outlet
weirs, and at the ends of the downcomer. Using modified downcomers that
decrease the downcomer volume helps to reduce the dead spot by keeping
more of the downcomer volume full of agitated liquid. For quick revamps
of existing equipment an alternative is to install metal or ceramic shapes to
modify the downcomer volume. Consider the influence of weir height and
avoid the use on inlet weirs when possible.

Baffle Trays

1. Shed Decks
Shed decks are essentially angle iron beams of various sizes from two to ten
inches that are placed in rows across the column. They typically are at on
24 inch tray spacing. They may be set in overlapping rows or rotated 90
degrees from tray to tray. The open area on the tray is typically 50%. The
positives of shed deck are that the fouling potential of the decks is almost
zero because there are no stagnate zones and low residence time.
Unfourtunaly, the efficiency of the tray almost matches the fouling
potential, particularly if wide shed decks are utilized. Shed decks work well
in fouling applications where their application is essentially for heat-
transfer purposes.
2. Disk and donut trays
Disk and donut trays are slightly sloped trays that allow the liquid to splash
from inner circle ring to outer circle ring. Fouling potential of this tray is
low along with the efficiency.
3. Side to side trays
Side-to-side trays are trays that allow the liquid to splash from side to side.
Fouling potential of this tray is low, as with efficiency.
Dual Flow Trays

Dual flow trays are the trays of preference for heavy fouling services. Dual
flow trays have no downcomers, where products of fouling phenomena can
accumulate or where polymer and solids can seed and propagate. Dual flow
trays are designed with enough open area on the tray decks to eliminate
stagnation and promote back mixing. A disadvantage to dual flow trays is
that they are limited in turndown potential.

The vapor and liquid transfer up and down the column through the holes
on the tray deck. This is an advantage in combating vapor-phase fouling as
the underneath of the tray is washed. The continuous agitation of the liquid
on the topside of the trays combined with continuous underside
wetting/washing action makes this tray suitable for fouling services.

The challenge of dual flow tray is mal-distribution in larger diameter

towers. The top of a column will move in a typical storm as much as six
inches. This movement will cause the hydrologic load to migrate in the
column. If hydrologic flow instability is developed it propagates down the
column. Improper feed, reflux or vapor distribution can also create
maldistribution problems,

Two types of dual flow trays are available; standard deck and rippled deck.
The standard deck has a flat plate, and the rippled deck has sinusoidal
waves.

Spray nozzles vs. distributors

There is much discussion in the industry between the advantages and

disadvantages of spray nozzles in fouling services. There are groups that
advocate only spray nozzles with no packing or shed decks. There are
groups that advocate only distributors with packing.

The best trough distributor will have superior distribution performance to

the best spray distributor. In fractionation services, the trough distributor
may be mandatory to meet separation requirements. In solids services,
such as refinery FCC slurry zones, specially designed trough distributors
have proven performance.

However, as long as the nozzles on a spray distributor have a large enough

minimum flow passage to protect against plugging, spray distributors resist
plugging. They have performed well in extremely severe services including
refinery vacuum wash zones and delayed coker main fractionators.

Material of Construction / Surface Finishing

It may be possible to reduce fouling by the choice of material selection, or

more importantly by surface finish. Smooth surfaces shed deposits easier
than rough surfaces.

A corroded surface is more prone to other fouling mechanisms. The

following guidelines on material selection can be considered on services
that are very prone to fouling.

1. Standard austenitic stainless steel such as 316 (18Cr 10Ni 3Mo) can be
utilized for resistance to acids and reasonable resistance to pitting
corrosion. Type 304L (18Cr 10Ni) stainless steel has a good resistance to
nitric acid. Austenitic stainless steels have relatively low strength, poor
anti erosion and abrasion properties and do not possess the ability to
resist stress corrosion cracking.
2. Super austenitic stainless steels with relatively high nickel content
(approximately 20 Cr 29 – 34 Ni), sometimes referred to as alloy 20 are
more costly than standard austenitic steels but provide excellent
resistance to acids and some acid chlorides.

3. Duplex stainless steel offer high strength, coupled with resistance to

abrasion, erosion and stress corrosion cracking.

4. Nickel based alloys, such as Hastelloy, have out standing corrosion

resistance in resisting acids, mixed acids, and acids at high
temperatures. The principle restriction in their use is the high cost.
 

Use of Additives

An additive may act to resist fouling in one or more of the following ways.

1. Chemically react with the fouling species to modify it’s fouling potential.
2. Change the physical interaction between the foulant and the equipment
surface.
3. Modify the deposit residing on the surface, so that is can be removed by
fluid flow.

It is also possible to reduce the fouling potential by pre-treating the metal

with passivating solutions.

May 3rd, 2012 in Design Distillation System

Operating Conditions affect on Fouling Phenomen
http://seperationtechnology.com/operating-conditions-affect-on-fouling-phenomen/

Over 25 Years Experience

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

Stagnant Zone

Stagnant zones are areas where fouling mechanisms can propagate:

vaporization of volatile components, polymerization, condensation,
sedimentation and chemical reactions. To eliminate stagnant zones a small
slope can be added to collection devices and flow promoters to distillation
trays. Stagnant zones increase the residence time of the process and lead to
precipitation and build-up of polymer and coke. Stagnant zones can also
promote unwanted chemical reactions because of high residence time and
little or no movement of material.

Sharp Transition

In the higher fouling cases sharp transitions and corners need to be

avoided. Transitions and corners are areas where polymer and solids can
seed and grow. Packed tower internals such as feed pipes and trough
distributors can be areas for polymer and solids build-up. The fouling
potential of packed tower internals can be reduced or eliminated through
proper design practices

If the fouling potential is in the vapor phase, the overhead vapor may be
drawn from the side of the tower versus the tower over head to eliminate
transition line length and an additional corner.
Emulsion Issue

In hydrocarbon service where water is present, as in ethylene quench and

saturator towers, emulsion issues can create opportunities for fouling. If
the pH of the water is allowed to depart from neutral ranges, emulsions can
occur that create mixtures that can carry hydrocarbons along.

Trace contaminates like oxygen, nitrogen compounds, sulphur compounds,

mercury, and chlorides have a significant effect on the performance of
chemical process fouling and this need to be recognized if present in the
process. Contaminates with surfactant properties should be reviewed
carefully.

April 10th, 2012 in Design Distillation System

Pinch Analysis : Part – 1
http://seperationtechnology.com/pinch-analysis-part-1/

Trillium Flow Technologies

Highly Engineered Pumps
Precision Engineered To Delivery Long-Term

Success As Your Partner In Flow Control.

trilliumflow.com

OPEN

Column Heat Utilization / Pinch Analysis

A distillation column has three basic sources of reject heat, the bottoms
product, the condensing overhead vapor, and the distillate product. The
two basic heat inputs are the reboiler and the feed.
Bottoms Product :

The ability to use the heat in the bottoms product will depend on its
requirements for downstream processing. If the product is required hot
downstream, it is impractical to cool it and then to reheat the bottoms
stream. If the stream does not need to remain hot, the following represent
possible uses of the bottoms liquid heat.

1. Preheating the column feed.

2. Use to run all or a portion of another column’s reboiler.

3. Exchange with another process liquid stream.
4. Steam generation and boiler feedwater heating.

Distillate Product – The options that apply to recovering heat apply

equally well to the distillate product. The opportunities of heat recovery
differ as the distillate product is at a lower temperature than the bottom,
and the distillate product may be a vapor, therefore containing a large
amount of heat in its vaporized condition.

Condenser Duty – The largest potential reject heat source of the distillation
column is the condenser. All this heat is available at essentially a single
temperature, and all the heat duty must be removed. Possible uses of the
condenser duty could be to supply heat to a neighboring column reboiler, to
produce waste steam, or to heat large liquid streams at low levels, such as
supplying hot water for a building.

Reboiler Duty – The reboiler represents the largest heat input to the
distillation column. The reboiler requires heat at a single high temperature.
It is desirable to minimize the steam consumption if possible by using
condensing vapors from other columns, hot process streams, or special
very low pressure steam.

Feed Preheating – The bottoms product or another hot liquid stream is

often used to preheat the feed. Whatever source used should cause the
maximum overall energy efficiently for the plant.

March 21st, 2012 in Design Distillation System

Pinch Analysis : Part – 2
http://seperationtechnology.com/pinch-analysis-part-2/

Trillium Flow Technologies

Highly Engineered Pumps
Precision Engineered To Delivery Long-Term

Success As Your Partner In Flow Control.

trilliumflow.com

OPEN

Changing the Columns Temperature

Two-Stage Condensation – For some multicomponent distillation columns

there is a broad range over which the overhead vapors condense (dew
point to bubble point). By using more than one condenser instead of a
single total condenser, we have the opportunity to recover some of the heat
at a higher temperature. For example, we could have two condensers, the
first condenser condensing part of the overhead to provide reflux, the
second condensing the distillate product. This situation is effectively a
partial condenser with vapors later condensed. The items to be emphasized
on a multi-stage condensation column are to avoid subcooling as much as
possible, and the recovery of the waste heat by steam generation in the high
temperature condenser.

Waste Heat Boilers – The use of a condenser as a waste heat boiler is

simple. The condenser is operated in a partially flooded situation, where
the level changes as the heat duty is changed. The water is boiled at
constant pressure in the tube side, and all steam produced sent to steam
headers or its ultimate use. The temperature at which the condenser
operates is important. For a temperature of less than 200°F, no sort of steam
can be produced. For higher temperatures the steam produced is
determined by the required pressure (1ATM, 40#, 150#, etc.). The condenser
will have to be larger as the temperature driving force goes down, so the
economics should be looked at. It may be necessary, in a case where low
pressure steam could be produced but no use exists at this low pressure,
although one does at a slightly higher pressure, to mechanically compress
the low pressure steam to a higher pressure, say from 25 psig to 40 psig.
Remember the true values given to the different steam pressures during the
evaluation of different waste heat steam generator possibilities.

April 12th, 2012 in Design Distillation System

Distillation Column Tray Selection & Sizing – 1
http://seperationtechnology.com/distillation-column-tray-selection-1/

Trillium Flow Technologies

Outstanding Flow Control
Our Products Are Backed By Expertise &

Support. Contact Us Today To Learn More.

trilliumflow.com

OPEN

Introduction:

Once the process design stage ends, the equipment design begins. This stage of
design converts process requirements into actual hardware.

One of the most prominent hardwares used for mass transfer is tray. Tray
columns are widely used in various types of mass transfer operations. All the
simulation results, which predict a certain number of theoretical stages, can be
converted to actual trays depending upon tray efficiency for a particular service.

In any conventional tray vapour rises through the liquid pool on the tray deck
and then disengages from the liquid in the space above the deck. Liquid enters
the tray from a downcomer above and leaves via a downcomer below.

Conventional Tray has three functional zones:

1. Active area for mixing vapour and liquid: This is the zone where mass
transfer occurs.
2. Vapour space above the active area: This is the zone in which liquid is
separated from vapour.

3. Downcomer between trays. This zone has two functions, first moving liquid
from one contacting tray to another and second disengaging vapour from
liquid.

Each of these zones takes up vertical and horizontal space in the tower.
Selection Guide for Tray Column:

The factors discussed below influence the choice between trays & packings. As
these are guidelines for selection of trays or packings for a particular service, it
is recommended to analyze each design case on its own merit for selection.

Sr.
No. System Favouring Tray Column System Favouring Packed Column

Solid Vaccum
1 handling system

High liquid
2 rate Low pressure drop application

Revamps-  
The pressure drop reduction
composition  
Feed and temperature
can 
3 be translated  into  capacity gain, an
variable
energy gain or separation
improvement.

Large diameter
4 columns Small diameter columns< 900 mm

5 Performance prediction is easy Corrosive system

Less weight saving in cost of

Foaming
6 system
foundations and
supports

Interboilers,  intercondensers,  cooling Low liquid holdup for reducing

7
polymerisation and
colils, & side draw
degradation.

High turn down Batch

8 requirements Distillation

9 Chemical reactions

The industry, based on its experience, has standardised the type to be used in
certain services. If this reference is not available the guideline as per Appendix
1 are to be used

Types of Tray

The particular tray selection and its design can materially affect the
performance of a given distillation, absorption, or stripping system. Each tray
should be designed so as to give as efficient a contact between the vapour and
liquid as possible, within reasonable economic limits.

 Valve tray:

Valve trays are perforated sheet metal decks on which round, liftable valves are
mounted. The vapour flows through valves which are installed parallel to the
outlet weir. Valve trays combine high capacity and excellent efficiency with a
wide operating range.

          Excellent liquid/ vapour contacting.

          Higher capacity.
          Higher flexibility than sieve trays.
          Can handle higher loadings.
          Low-pressure drop than bubble cap.

 
 Sieve tray:

Sieve trays are flat perforated plate in which vapour rises through small holes in
tray floor, & bubbles through liquid in fairly uniform manner. They have
comparable capacity as valve trays.

       Advantages:

       Simple construction Low entrainment,

       low cost Low maintenance cost
       Low fouling tendency

       Disadvantages:

                                                                   Less-flexible to varying loads than

                                                                   other two types
 Bubble cap tray:

Vapour rises through risers or uptakes into bubble cap, out through slots as
bubbles into surrounding liquid on tray. It is mainly used in special applications.

Advantages:

 Moderate capacity
Most flexible (high & low vap. & liquid rates)
Can provide excellent turndown.

Disadvantages:

High entrainment,                                                                                            High

fouling tendency
  High cost,
  High pressure drop

Dual flow trays:

A dual flow tray is a sieve tray with no downcomers. This tray operates with
liquid continuously weeping through the holes. Due to the absence of
downcomers, dual flow tray gives more tray area hence a greater capacity than
any of the common tray types. They are ideal for revamp where if some
efficiency can be sacrificed for more capacity. They are least expensive to make
and easiest to install and maintain.

 Dual Flow Tray                                                                               Baffle Tray

 Baffle trays:

For a baffle tray column the gas flows upwards through the baffle openings and
in doing so contacts the liquid showering down from one baffle to the next.
Baffle tray columns have almost same flooding capacity as cross flow trays.
Types of baffles used are disc & donut and segmental baffles for various column
diameters.

Dual flow and baffle trays are used for fouling applications, solid / slurry
handling services, corrosive services.

Proprietary types of trays:

MD Trays – Linde / UOP,

Ripple Trays – Stone & Webster Engg. Corp.
Rectangular Valve (BDH),
ValveGrid (MVG/SVG),
SHELL HIFI, ConSep Trays – SulzerBallast Tray,
Flexitray, Bi-FRAC, SUPERFRAC and ULTRAFRAC Trays – Koch-Glitsch Engg.Co.,
Tunnel Trays- Montz,
Nye trays- Nye Engg Co,
Comparison between Common Conventional Trays.

Sr. Bubble-Cap
Factors Sieve Tray Valve Tray Tray Dual-Flow Tray
No.

Moderately
1 Capacity High High High Very High

Moderately
2 Efficiency High High High Least

3 Turndown ~50% ~25-30% 10% Least

4 Entrainment Moderate Moderate High Low to moderate

Pressure
5 Drop Moderate Moderate High Low to Moderate

~ 2-3 times of
~1.2 times sieve
6 Cost Low Least
sieve trays trays

Low to
7 Maintenance Low Relatively High Low
Moderate

High: Tends to
Fouling Low to collect
8 Low Extremely Low
Tendency Moderate Solids

Effects of Low to
9 Low High Very Low
Corrosion moderate

Some information
Proprietary,
Design Available.Instability
10 Well Known but readily Well Known
information can occur in large
available
dia. (>8 feet)

Extremely low
Liquid
Often used Where high Capacity revamps,
Main flow & Where
11 when turndown Highly fouling and
turndown is
Application leakage must be
 is not
critical required corrosive services
minimized

Tray Parameters

 a)     No. of passes (Np):

The numbers of flowpaths of liquid on tray are 1, 2, 3 or 4 as per liquid capacity

requirement of column. From a capacity viewpoint, a liquid rate greater than 6
gpm / inch of weir (weir loading), is the rate at which a higher number of flow
paths should be considered. The maximum allowable weir loading is 13 gpm/in
of weir length. If the weir loading exceeds this the tray needs redesign with
higher number of passes.

 b)     Tray Spacing (S):

 Tray spacing is the distance between two trays. Generally tray spacing ranges
from 8 to 36 inches (200 mm to 900 mm). Prime factor in setting tray spacing is
the economic trade-off between column height and column diameter. Most
columns have 600 mm tray spacing. Cryogenic columns have tray spacing of
200-300 mm. 

c)      Outlet Weirs (hw):

 An outlet weir maintains a desired liquid level on the tray. As the liquid leaves
the contacting area of the tray, it flows over the tray weir to enter into the
downcomer.

 d)     Downcomer Clearance (hcl):

This is the vertical distance between the tray floor and the bottom edge of the
downcomer apron. The Normalpractice is to use a downcomer clearance of 1/2
inch less than the overflow weir height to provide a static liquid seal

 e)     Inlet Weirs & Recessed Seal Pans:

 Inlet weirs and recessed seal pans are primarily used for achieving a
downcomer seal in cases where a potential positive sealing problem exists and
clearance under downcomer is limited
f)       Downcomers:

 Passage of liquid from the top tray to the bottom of tray occurs via
downcomers. Downcomers are conduits having circular, segmental, or
rectangular cross sections that convey liquid from upper tray to a lower tray in
a distillation column.

 g)     Downcomer width (Chord height, WDC):

It is maximum horizontal distance between tower wall and weir.

 h)     Flow path length (FPL):

 Flow path length is the distance between the inlet downcomer & outlet
downcomer. The minimum limit for flow path length is 400 mm in order to
provide good contacting between vapour and liquid. This is also necessary for
the mechanical reason of providing tray manway.

 i)       Tray deck thickness (t):

 Trays normally used in commercial service need a minimum material thickness

to provide structural strength (personnel walk on them during installation) and
corrosion allowance. A thickness of 10 to 12 gauge (2.5 to 3.5 mm) is customary
for carbon steel, while 12 to 14 gauge (1.9 to 2.5 mm) is used for stainless steel
trays (in general no C.A. for SS)

 j)       Hole pitch (P):

 Centre to centre distance between holes is called pitch. Normal practice is to use
a hole pitch to hole diameter ratio between 2.2 to 3.8.

 k)      System (Derating) factors:

 Derating factors are often closely related to the foaming tendency of the system.
Higher the foaming tendency, the lower is the Derating factor. System factors
are used in three of the rating correlations (jet flood, down comer backup flood,
down comer choke) to account for system effects on hydraulic capacity limits. It
includes both foaming effects and high vapour density.

 l)       Bubbling (Active) Area (AB):

Bubbling area is the column area, which is actually available for vapour
bubbling through liquid. It can be defined as column area minus downcomer
areas, downcomer seal & large calming zones.

 m)   % Hole Area:

 This is the ratio of hole area to bubbling area. The default practice is to target a
hole area of 8 to 10 % of bubbling area for pressure services. The acceptable
range for percentage hole area is 5 % to 15 %. However for some critical
services, we can go % hole area up to 17-17.5% provided that weeping is under
control. Hole areas below 5 % are not used.

 n)        Anti jump baffles:

Anti jump baffles plates suspended vertically above centre or off centre
downcomers, which stops liquid jumping from one deck onto the opposite deck,
flow path

Tray Hydraulic Parameters

Following are the some important output parameters of tray hydraulics.

 a)     Flood:

Jet Flood:

In spray regime operation flooding is brought about by excessive vapour flow,

causing excessive liquid to be entrained in the vapour up the column. In froth
and emulsion flows regimes operation excessive froth entrainment in the
vapour up the column causes jet flooding.
Down-comer Back-up Flood:

Occurs when the pressure available for a given height of liquid and froth in the
downcomer cannot overcome the total pressure drop across the tray This
pressure imbalance causes the froth in the downcomer to start backing-up until
it reaches the tray above, causing an increased accumulation of liquid on it. It
requires high liquid and vapour loads.

Downcomer Choke Flood:

The mechanism by which this type of flooding occurs is one related to frictional
pressure losses in the downcomer becoming excessive. In addition, the vapour
carried into the downcomer must separate from the liquid and then flow
counter-current to the liquid entering the downcomer. When the combination
of vapour exiting and the liquid entering becomes excessive, the downcomer
entrance is choked causing the liquid to backup on the tray. It requires relatively
high liquid rates, surpassing a velocity limitation on the downcomer.

b)    Weeping/Dumping

The pressure exerted by the vapour is insufficient to hold up the liquid on the
tray. Therefore, liquid starts to leak through perforations.

c)    Pressure Drop:

Pressure drop is an important consideration while designing a tray. It becomes

more critical for the vacuum systems than the high-pressure systems. The tray
pressure drop is viewed as the sum of the pressure drop through the valves or
sieves and pressure drop through the aerated liquid on the tray deck.

d)    Turndown ratio:

Turndown ratio defines the range of vapour load between which the column
can operate without substantially affecting its’ primary separation objective (i.e.
fractionation efficiency) or over which acceptable tray performance is achieved.
The tray efficiency stays at or above the design value throughout the turndown
range.

Tray Sizing

 The sizing procedure is an iterative calculation. A preliminary design is set, and

then refined by checking against the performance correlations until an
adequate design is achieved. The sizing calculations are performed at the point
where column loading is expected to be highest and lowest for each section, i.e.,

i)  The top tray

ii) Above every feed, product drawoff, or point of heat addition or removal.

iii) Below every feed, product drawoff, or point of heat addition or removal.

iv)  The bottom tray.

v)   At any point in the column where the calculated vapour or liquid loading
peaks

The sizing is done at all above load points and also detailed sizing is checked at
all above load points. All design parameters given in the design procedure
below are calculated at all above load points at turndown and turn-up loads so
that the feasibility of design for varied loads is tested.

a) Preliminary determination of tower area:

The methods used for determining tower diameter are:

“C” Factor Method

Nomograph Method

FRI Tray design handbook

However in this technical guideline we are describing method using C-Factor

Method.

C-Factor Method:

The following calculations are done at all the loading points mentioned above
and diameters are found separately. If the difference in calculated diameter at
different sections exceeds 20 percent, different diameters for the sections are
likely to be economical. The section having different diameter should be at least
20ft in length else same diameter can be maintained.

i.   Tray Area

Assume appropriate values for following parameters (based on system
requirements) for preliminary diameter calculation.

dH = Hole diameter, inches (¼ to ½ inch) S = Tray spacing, inches (18 – 24″)

hct = Clear Liquid height at the transition from the froth to spray regime, in of
liquid.

Assumption: The starting values for these can be dH=1/4″, S=24″, h ct=2″

Calculate C-Factor (CSB) using following Kister and Haas Correlation:

ii. Flood Velocity Calculation

This is the velocity of upward vapour at which liquid droplets are suspended.
Calculate Flood Velocity (uN) using following equation:
iii. Net Area Calculation 

The net area represents smallest area available for vapour flow in the inter-tray
spacing. Calculate Net Area (AN) from the flood velocity using following
equation: Assume the column is to be designed for 80% of flood.

iv. Downcomer Area Calculation Calculate downcomer area (AD) from clear

liquid velocity in downcomer using following formula:
Where,

QL       = Liquid Flow Rate, ft3/s

VCL     = Clear Liquid Velocity in Downcomer

Value of VCL obtained from table below. No derating factor is required for this
calculation, as VCL values have taken care of foaming

Table: Recommended VCL values for different foaming tendencies

VCL  in downcomer, ft/s

Foaming
Example 18-in 24-in 30-in
Tendency
Spacing Spacing Spacing

Low pressure (<100-psia) light

Low hydrocarbon, 0.4-0.5 0.5-0.6 0.5-0.6

stabilizers, air-water simulators

Oil systems, crude oil distillation,

absorbers,
Medium 0.3-0.4 0.4-0.5 0.4-0.5
med. pressure (100-300 psia)
 hydrocarbon

Amine, glycerine, glycols, high-

pressure 0.2-
High 0.2-0.25 0.2-0.3
(>300-psi) light hydrocarbons 0.25

v.   Tower Diameter Calculation

 TotalTowerArea (AT) = AD  + AN

b)         Preliminary tray layout:

 A Preliminary layout is needed as layout influences the column size.

Downcomer Layout:

Check the % of Downcomer area with respect to tower area:

The Fractional area should around 10% but avoid less than 8% in normal
circumstances. Note that AD should in no circumstance be less that 5% of AT

Net Area (AN):

The total tower cross-section area AT less the area at the top of the downcomer
(sometime refer to as free area, the term free area.)

The net area represents the smallest area available for vapour flow in the inter-
tray spacing.

AN    = AT    – AD

Bubbling (Active) area (AB):

The total cross-section area AT less the area at the inlet & outlet downcomer is
called as bubbling area.

 AB    = AT    – ADT   – ADB

 Below figure shows the Typical Tray Layout.

Weir Length and Downcomer Width: SinglePass Tray:

The calculation of Weir Length and Downcomer Width involves geometrical

relationship between downcomer area, downcomer width, and downcomer
length.

Following Figure shows downcomer geometry:

Calculate downcomer width and weir length using following method

? = sin-1(h/R)

w = 2*R COS (?) or w = 2*(R2 – h2)0.5

?/2 = ?/2 – ?

Sector area = ASECT  = ? R2 * ? / (2 * ?)

Area of triangle (ABC) = ATRI.  = w*h/2

Where,

Lw  = Weir Length = w* (1-fractional weir blockage)

wdc  = Downcomer Width = R -h

AD  = Adc  = Downcomer Area

Fractional weir blockage is the fraction of total weir length that is available for
liquid flow by using picket and fence type of weir. Blocked (Picket fence) weirs
are used for handling low liquid loading.

Down-comer area

AD = ASECT– ATRI

Two Pass Tray:

Two pass trays have alternating arrangements of one center-downcomer and

two side-downcomers.

The side downcomer area can be calculated as that for single pass tray. It should
be noted that side down-comers are on both sides.

Center downcomer calculations can be done as follows in similar manner as

side down-comer:
? = sin-1  (h/R)

w = 2*R COS (?1) or w = 2*(R2 – h2)0.5

? = 2*(?/2- ?)

Sector area = ASECT = ? R2 * ? / (2 * ?)

Area of center downcomer = Area of circle -2*area of sector + 2*Area of Triangle

Area of downcomer = ?*R2 – 2* ASECT + h1*w1

In case of more than two pass trays we have to define one more parameter, i.e.
off-center downcomer location from centerline. This needs to be done on a case-
by-case basis.

Liquid Flow Path Length (FPL):

ForSinglePassTray:

FPL= (tray diameter) minus (side DC width of the tray) minus (bottom width of
DC of tray above)

 
 

Where,

Downcomer width (Centre downcomer, Bottom

w1dc = of Downcomer)

Downcomer width (Side downcomer, Top of

w2dc = Downcomer)

Downcomer width (Centre downcomer, Top of

w3dc = Downcomer)

Downcomer width (Side downcomer, Bottom of

w4dc = Downcomer)

C) Detailed Design

Flooding Check:

The flooding check is performed using following Correlations:

1. Kister and Haas correlation.

2. Downcomer choke-Koch correlation
3. Fair’s correlation

4. Smith et al. correlation

1.  Jet Flood: Kister and Haas correlation

This correlation possess following advantage:

–         It gives a close approximation to the effects of physical properties,

operating variable, and tray geometry on the flood point.

–         It describes spray regime entrainment.

–         It was derived from a much wider database of commercial and pilot-scale
column data.
–         It can predict sieve and valve tray entrainment flooding within ± 15 and ±
20 percent respectively.

This correlation possess following restriction:

Sr.no. Factors Applicability

Flooding Entrainment (Jet) flood

1 Mechanism only

2 Tray Type Sieve or Valve trays only

3 Pressure 1.5-500 psia

4 Gas Velocity 1.5-13 ft/s

0.5-12 gpm/in of outlet

5 Liquid Load weir

6 Gas Density 0.03-10 lb/ft3

7 Liquid Density 20-75 lb/ft3

8 Surface Tension 5-80 dyne/cm

9 Liquid Viscosity 0.05-2.0 cP

10 Tray Spacing 14-36 in

11 Hole Diameter 1/8-1 in

Fractional Hole
12 Area 0.06-0.20

13 Weir Height 0-3 in

Steps to calculate % Flooding using Kister and Haas correlation:

i. Calculate Weir Load (QL):

Liquid Load describes the flux of liquid across the tray.

 

ii. Clear Liquid height at the transition from the froth to spray ((hct)
  

2. Jet Flood: Fair’s correlation

The Fair correlation has been standard of the industry for entrainment flood
prediction. Fair’s correlation tends to be conservative, especially at high
pressure and liquid rate.

This correlation possess following restriction:

 Sr.No Factors Applicability

1 Flooding Mechanism Entrainment (Jet) flood only

Sieve Tray, Valve and Bubble-cap

2 Tray Type Tray

3 Hole size
Hole£ ½ in (sieve tray)

Weir height < 15% Tray Spacing

Steps to calculate % Flood using Fair’s correlation:

i. Calculate flow parameter

 
 

3. Down-comer choke-Koch correlation:

This is the more conservative correlation for checking Down-comer Design.

Steps to calculate % Load Utilization using Kister and Haas correlation:
 

4. Hydraulic checks

Hydraulic check involves checking following parameters:

–    Flow Regime

–    Entrainment

–    Downcomer residence time

–    Pressure Drop

–    Downcomer backup

ii.Determination of Flow Regime

Froth Regime

This is the most commonly encountered flow regime in operating columns. The
froth formed under this regime is described as one where the size and shape of
bubbles is non-uniform and with rather large size distribution, as well as
travelling at varying velocities. The liquid surface is either wavy or it presents
oscillations. This is a liquid continuous flow regime.

Spray Regime
This regimes occurs at relatively high vapour velocities (i.e. large vapour flow
rates) and low liquid loads, characteristics which are typical of vacuum systems.
The vapour velocity is so large, that the liquid phase is completely disrupted and
is no longer a continuous phase on top of the tray; liquid is a dispersed phase
present only in the form of drops, and therefore the continuous phase is the
vapour.

Emulsion Regime

This flow regime is typically encountered in high-pressure systems and

relatively high liquid loads. The shearing action of the high velocity liquid “tears
off” the vapour bubbles leaving the orifices on the tray. Most of the gas is
emulsified in small bubbles within the liquid, with the mixture behaving as a
uniform two-phase fluid, obeying the Francis weir formula. This is a liquid
continuous flow regime.

The determination of regime on tray given below is only for information and
has no use in sizing.

ii. Froth-Emulsion Transition Check

This correlation is applicable for Sieve trays only.

The value of actual flow parameter is calculated as below:

 
 

If the value of actual flow parameter exceeds 0.0208 then the regime of
operation is emulsion.

iii. Froth-Spray Transition Check:

Porter and Jenkins correlation for the froth to spray transition.

Where,

Lw – weir length in inches, AB – Active area ft2

p – pitch in inches

hc  – clear liquid height, inches

5. Entrainment:
If entrainment is excessive, column diameter or tray spacing are usually
increased. As recommended value, the entrainment from the tray should not
exceed about 0.10 lb liquid entrained per pound of liquid flow.

Methods to determine Entrainment:

Fair’s entrainment correlation

This method holds good for froth and emulsion regime. However it is less
accurate for spray regime. For a trays operating at a high liquid to vapour ratio,
0.1 lb of liquid entrained per pound of liquid is an excessive quantity of
entrained liquid.

Kister and Haas Correlation

This method is used for Spray Regime; Es is entrainment lb of liquid / lb of

vapour.
 

May 7th, 2012 in Design Distillation System

REDUCING HEAT LOSSES USING INSULATION
http://seperationtechnology.com/reducing-heat-losses-using-insulation/

Trillium Flow Technologies

Application-Specific Pumps
Complete Project Support Through Each

Stage Of The Life Of Your Operation.

Contact Us Now.

trilliumflow.com

OPEN

A different situation could occur for a column that operates at 400°F

bottom, 300°F top, using 600 psig steam in the reboiler, producing 25 psig
steam in the condenser. For this case, insulating the column bottoms and
reboiler will save valuable 600 psig steam so much is needed. Insulation on
the upper section of the column is also valuable as saved heat generates
useful 25 psig steam in the condenser. So this entire column needs
insulation.

In the case of a column operating in the refrigerated condition, insulation

must be used on the condenser and top portion of the column to prevent
heat flowing into the column, which would then have to be removed by
expensive refrigeration. If the reboiler and bottom sections of the column
are also cold insulation will also be required as these sections are part of
the coolant cycle.
June 18th, 2012 in Operations of Distillation System
Column Revamping
http://seperationtechnology.com/column-revamping/

Over 25 Years Experience

Horizontal & vertical coalescing filters, dual

closure coalescing filters, dust filters.

Sepra-Chem Corporation

 
 

Additional or More Efficient Trays

According to Mix, et al, tray changes are economically feasible if ,

          . . . . . . . (1)

N = Number of trays in the column

? = Murphie Plate efficiency

a = Relative volatility (light to heavy)

S = Separation Factor
D = Distillate

B = Bottoms

LK = Light Key

HK = Heavy Key

x = Concentration, mole fraction

P = Column pressure in ATM

K = Reflux ratio R/RM

Before one proceeds, it is recommended that a rigorous distillation

calculation be performed on the existing column using the actual
temperature, pressures, compositions, etc. of the column. Distillation
programs that have been developed by Chemshare, Simulation Sciences,
Phillips 66, and others for simulating your column are available through
various computing service bureaus.

A plot can be made of the distribution of the various components tray by

tray. This plot may indicate the feed tray may be changed or additional
trays may be beneficial if entirely in the rectification or stripping section of
the column.

If equation (1) shows the column may benefit from more trays, you can run
several cases with reflux as the variable (heat load changes) and determine
the saving in energy.
You have the option of adding more trays or replacing existing trays with
more efficient type trays.

Additional Column Draw

Three possible column draw options are pasteurization, intermediate
product and intermediate impurity. Pasteurization means the removal of
light ends from the distillate by venting off the accumulator and removing
the distillate product several trays below the top. Six criteria are listed by
Mix, et al.

The intermediate product is considered when the temperature difference

between bottom and top exceeds l00ºF, and when the split of one key
between two products is desired. The intermediate impurity drawoff is
useful for removing impurity buildup under high reflux operations. The
impurity flow rate must be less than .01 times the feed rate in lb moles per
hr, and the relative volatility between the light and heavy key less than 1.5.

June 21st, 2012 in Design Distillation System

Trillium Flow Technologies
Application-Specific Pumps
Complete Project Support Through Each

Stage Of The Life Of Your Operation.

Contact Us Now.

trilliumflow.com

OPEN
A computer program has been developed to calculate the effective liquid
heating and vapor cooling duties as functions of temperature throughout a
simulated distillation column. The heating and cooling curves for the base
case Demethaniercolunm from Figure 1 have been calculated using this
program, and the results are shown in Figure 3. About 170 MMBtu/hr are
transferred in this column, reboiler, and condenser as the liquid is heated
to 54OF while the vapor is cooled to -139?F.

For maximum energy and capital efficiency, the heating and cooling curves
should be approximately parallel but slightly closer at lower temperatures.
In Figure 3, the heating and cooling curves approach each other near the
bottom of the column at 40OF where methane is almost completely
separated from ethylene. Below this point heat is exchanged primarily
between ethylene and propylene in the vapor and liquid streams. The
heating and cooling curves also approach each other just above the bottom
feed at -20OF, just above the middle feed at -94OF, and just below the
condenser at -135OF. This column is well designed to optimize the thermal
efficiency given the limitations of three feeds, a single interexchanger, and
45 theoretical stages. Figure 4 shows the liquid mole percents of methane,
ethylene, ethane, and propylene throughout the column as functions of
temperature. This figure emphasizes the separation between methane and
ethylene taking place with propylene entering the column with the middle
and bottom feeds at -94OF and -20OF. At the very bottom of the column,
propylene is condensed by vaporizing ethylene. This occurs because some
propylene is vaporized in the reboiler along with the methane and
ethylene, and it must then be condensed further up the column. The same
phenomenon occurs at -40OF just above the bottom feed, and contributes to
the thermodynamic inefficiency of the process design.
June 8th, 2012 in Design Distillation System, Distillation Technology,
Operations of Distillation System
Optimization of Distillation System Operations
http://seperationtechnology.com/optimization-of-distillation-system-operations/

Water System Installation

Your MMS Water Filter Shop
MMS drops, CDS/CDL and CDSplus chlorine

dioxide products. Buy our top products now

aquarius-prolife.com

OPEN

Your operating procedures were probably written before the large increase
in energy cost drew attention to energy conservation as one primary
objective. In addition, the operators are probably using the procedures only
as a guide and have developed their own procedures based upon ease of
operation.

Reducing the Reflux Ratio of Columns:

The optimization of the reflux ratio of the distillation column can produce
significant energy savings. The investigation can start by checking the
operating manual and column performance specifications for the design
conditions, including the reflux ratio. If the design conditions are no longer
valid due to changes in feed composition or product requirements, it is
recommended that a vigorous distillation calculation be made. If the
calculations are very difficult, you can make use of commercial computer
programs made available through various computing service bureaus. The
design reflux should be compared with the actual ratios controlled by each
shift operator. The daily laboratory analyses of the column products are
compiled and compared with the design specifications. If the column is
operated at a reduced production rate, the design reflux rate should be
calculated for this reduced rate.
It is extremely difficult to change people, even more difficult when it
requires more work effort without visually seeing the results. If one
operator was found who operated the column at a lower reflux ratio than
the others, you might get the confidence of the operators by getting all the
operators to maintain this ratio. If you merely write a note in the unit’s
operating log leaving instructions, you will probably not be successful in
lowering the reflux ratio. You must work closely with the superintendent,
foremen, and operators instilling confidence as you show the energy
savings resulting from their efforts. If the operating department has
monthly meetings for the supervisory people, you can use it as a forum to
present your objectives, how you plan to approach them, and request their
support and assistance. Later you can report progress and discuss problem
areas.

Steam or fuel usage per pound of product can be tabulated daily along with
reflux ratio, product purity, etc. and compared with column performance
before the change. The savings in energy can be converted to a monetary
value and reported to the operating people. As an alternate you might
represent the energy savings as barrels of imported oil per year.

As the reflux ratio is reduced, a point will be reached at which the

operators are overworked and having difficulty in maintaining product
purity. This is the opportunity to show your concern to the operators by
backing off on the ratio.

Lowering Product Specifications

Sometimes, product specifications can be lowered. Who decided on the

present product specifications? Are they justifiable? For example, the sales
group may have had the product purity increased to justify selling more
product and beating the competitors. The buyer may require a purity in
excess of his real needs. Higher purity product requires more energy to be
consumed per pound of product. Since the sales department has probably
expressed an optimistic opinion as to the value of higher product
specifications in the market place, an economic analysis based upon their
opinions would most likely say to make no specification changes. A better
approach may be to analyze the specification requirements for each type of
user of the chemicals and determine if the higher specification is required.
A different selling  technique  may  retain  the  customer  even  if  product 
specifications  are lowered to save energy.

If the product from the column is feed to another unit in the plant, then the
effect of lowering the purity on the other unit must be determined. Thus,
the energy conservation project requires the additional collection and
tabulation of operating data. A statistical approach may be required to fully
interpret the results of changes due to the variability of the processes by
changes in other parameters.

Lowering Steam Usage

One of the most talked about energy wasters is steam leakage from “bad”
steam traps and leaking fittings. Steam traps are blamed for being
inefficient or worn out and causing as much as 10% of the generated heat
from steam to be lost. Is this true or just a sales method to sell more traps?
It turns out that steam leaks cause a significant energy loss.

Mr. Goyette, in his article “Estimating the Costs of Steam Leaks”, (see
appendix 7-C) shows the cost effect of steam leakage from various size holes
(1/8″, 1/16″, and 1/32″) in a 150 psig steam system. The cost was based upon
incremented steam costs. An example showed that a 1-inch union was
found leaking at a loss of $3000 per year. The repair cost was $50 or a six
day payout. Of all the energy savings steps that the Tenneco plant did, Mr
Goyette said the single largest contributor was steam-leak repairs. Steam
traps will wear out. Armstrong Machine Works claim that the inexpensive
disk type steam trap wears out in 6 months and should be replaced that
frequently. If condensate is recovered, leaking traps can cause an excessive
return temperature and cause failure of the condensate return pumps.
Severe water hammer can occur as hot steam contacts condensate that has
cooled below the temperature of the steam. The following steps are
recommended for saving energy in your steam condensate distribution
system and starting an effective steam energy management program:

Develop an estimate of the cost of steam leaks based upon your plant
costs similar to the Goyette article described above. A method for
 demonstrating visually to plant people what these losses are can be
made.
Run a survey, recording all leaks, size, cost, and location.

Check the operation of all installed disc traps used for drips and steam
tracing. If found leaking, consider replacing with a more efficient type
trap. Before replacing, check installation design and confirm trap size
(not over or undersized).
Check installation and operation of steam traps used on equipment
using the sound detection method, the pyrometer method, or the glove
method. The installation should be checked for proper trapping. Items
checked include strainer, check valve, back pressure, orifice, and inert
gas venting. Improper venting can cause a severe reduction in heat
transfer rate.

Check vent valves on steam jacketed equipment and kettles for proper
operation (removal of inerts without steam loss).

Start a preventative maintenance program to maintain the steam

distribution system in excellent condition. If manpower is not available
in maintenance, you can have the operating people maintain a simple
log for their area of responsibility.
Steam trap manufacturers will be happy to furnish information to assist
in your energy saving program to reduce steam losses, but use your own
economic costs to decide whether to replace, repair, or redesign the
system.

There is insufficient published information to say that 10% of the steam is

wasted by steam traps, but some major chemical companies have invested
large amounts of manpower and money to replace or revise steam trapping
systems in their plants.

April 30th, 2012 in Operations of Distillation System

Gas Liquid Verticle Separator
http://seperationtechnology.com/gas-liquid-verticle-separator/

Over 25 Years Experience

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

This procedure guideline covers sizing of Vapor/Liquid Vertical Separator.

In a vertical separator fluids enter the vessel through an inlet device whose
primary objectives are to achieve efficient bulk separation of liquid from he
gas and to improve flow distribution of both phases through the separator.
Liquid removed by the inlet device is diverted to the bottom of the vessel

while the gas moves upward, usually passes through the mist extractor to
remove small droplet of entrained liquid, and then the vapour phase flows
out of vessel.

As a general rule a vertical separator is chosen when the ratio of vapour to

liquid volume is large i.e. 750 or more.
The various type internal available for
vertical separator, these are:
1. Demister Mats(Mist Extractor)

2. Vane Packs

or Swirl Decks.
3. Mutilcyclone
4. Filter Separators

5. Cyclones

There are separator where inlet diverter is can also be used(As Per GPSA).

Sizing of Vertical Separator without Mist Extractor(Demister):

and liquid
An empty knock-out drum is suitable for bulk separation of gas
especially when the liquid volume fraction is high stratified or plug flow in
the inlet pipe It should also be used when internals are to be kept to a
minimum eg. Fouling servics (wax, coke, scale etc..)

normal operating gas
The design margin of 10-25% should be applied to the
flow to calculate the free cross section area above the high level trip.

Determination of Separator Diameter:

·  Based on K-Value (Using Souder – Brown Equation):

The Design of separator is based on Souders-Brown Equation:

Vmax = Maximum velocity (Terminal Velocity) gas velocity allowable for

particles of size Dp to drop or settle out of gas , m/s

?L = Vapor Density, kg/m3

?V = Liquid Density, kg/m3

K = Empirical Constant for Sizing of Separator, m/s

g = Acceleration due to gravity, m/s2

Dp = Diameter of Particle, m

For Process Separator.

micron.
Size Range is 150 to 300
For Flare Knock Out Drum

Size Range is 300 to 600 micron.

For conservative calulation use smallest size in the range.

C’ = Drag Coefficient.

This obtained from the

graph of C'(Re)2 versus Drag Coefficient (C’)
Where,

Where,

?v = Density of Vapor, lb/ft 3.

?L = Density of Liquid, lb/ft 3.

Dp =Diamter of Particle , ft.

? = Viscosity ,cP.

The above formula

is in FPS Unit.

While in MKS Unit the above formula is

Where,

?v = Density of Vapor, Kg/m3.

?L = Density of Liquid, Kg/m3.

Dp =Diamter of Particle , micron.

? = Viscosity ,cP.

After calculating
value of C'(Re)2 value of C’ (Drag Coefficient) can be
obtained from the below graph.
The Vessel diameter calculated shall also satisfy the following Criteria:
If Liquid De -Gassing is Required:

Where,

Flow Rate , m3/sec.
Q = Liquid
? = Dynamic Viscocity , Kg/ms.

?v = Density of Vapor, Kg/m3.

?L = Density of Liquid, Kg/m3.

If Liquid De -Foaming is Required:

June 15th, 2012 in Vapor Liquid Equilibria
Dividing Wall Distillation
http://seperationtechnology.com/dividing-wall-distillation/

Trillium Flow Technologies

Outstanding Flow Control
Our Products Are Backed By Expertise &

Support. Contact Us Today To Learn More.

trilliumflow.com

OPEN


Butadiene Technology Overview

The BASF butadiene technology uses NMP as solvent. NMP has unique
properties that make it an excellent solvent for this service. NMP is the least
toxic and most stable of the solvents used in butadiene extraction.

The BASF butadiene extraction process consists of three major sections

(Figure 3). The first section, extractive distillation, is used to separate the
compounds that have relative volatilities close to butadiene. Butanes and
butenes are separated in the first extractive distillation, while C4 acetylenes
are separated in the second extractive distillation. The second section of the
butadiene plant is conventional distillation. In this section, methyl
acetylene, which is more volatile than 1,3 butadiene is separated. Also,
compounds less volatile than 1,3 butadiene, such as 1,2 butadiene, ethyl
acetylene, and C5 compounds are separated by distillation. The 1,3
butadiene product meeting the required specification is produced in this
section.
The third major section of the plant is solvent degassing. In this section the
C4s are completely separated from the solvent. Lean solvent and most of
the C4s are returned separately to the extractive distillation section of the
plant. A small stream containing the C4 acetylenes is separated in the
degassing section.

Dividing Wall Column Technology

Figure 4 shows a schematic of the “classic” dividing wall distillation
scheme. The feed consists of components A, B and C, and the objective is to
separate component B from components A and C. This is the same objective
as in the distillation section in a butadiene extraction plant. The dividing
wall distillation scheme has the following advantages and disadvantages:

Advantages                                                            Disadvantages

Lower energy consumption                           Taller distillation column

Less investment                                                  Single operating pressure

Less equipment/less maintenance             Higher pressure drop

Less thermal stress                                           Higher temperature difference

Lower energy consumption

Lower energy consumption can be achieved in a dividing wall column

system mostly due to a reduction in energy loss associated with entropy of
mixing. With one feed point, the dividing wall column system can be
designed to minimize the mixing entropy loss compared to the two-column
system. There is also a small savings in heat loss with a single dividing wall
column compared to the two-column system. However, since the BASF
butadiene process already has an efficient heat recovery system, the overall
gain in energy efficiency by using the dividing wall column system can vary
depending on the specific design. A corresponding investment cost savings
related to using less energy in the dividing wall column system is also
possible. The investment cost savings is based on simplifying the heat
recovery system and thereby eliminating equipment.

Less investment
Less investment cost in the column system is the main advantage of the
dividing wall column design. Figure 5 shows the current configuration of
the extractive distillation section with three columns: the Main Washer, the
Rectifier, and the Afterwasher. Figure 6 shows the same system using the
dividing wall configuration that combines the Rectifier and Afterwasher
into one column. The result is one less column, and elimination of one pair
of pumps used to pump the Afterwasher bottoms to the Rectifier in the
conventional scheme.


Figure 7 shows the conventional scheme for the distillation section of the
butadiene plant. The first column is used to separate propyne from the 1,3
butadiene, while the second column separates the 1,2 butadiene and
heavier compounds from 1,3 butadiene. Using a dividing wall column, these
two column systems can be combined into one column system shown in
Figure 8, resulting in a significant reduction in piece count and investment
cost. The equipment eliminated is one column, one overhead system
(condenser, drum and pumps), and one reboiler. The total reduction is six
pieces of equipment.

The above two systems combined result in nine less pieces of equipment.
Since a typical butadiene unit contains 90 to 100 pieces, this piece count
reduction gives an overall savings of about ten percent in engineering costs
and five to eight percent in equipment and commodities costs. Construction
costs will also be reduced correspondingly resulting in an overall
investment savings of five to eight percent.

Less Equipment/Less Maintenance

Since nine pieces of equipment are eliminated using the dividing wall
column technology, there is also a savings in maintenance costs associated
with the plant. In particular, two pump services (four pumps) are
eliminated. This reduces the costs associated with maintaining these pumps
(seal repairs, etc.).

Less Thermal Stress

Although it is not a major consideration for butadiene extraction plants, the

combination of column systems into dividing wall column systems does
reduce the heat losses and thermal stress in the plant. Combining two
columns into one means considerably less piping, and less heat loss from
the columns. Since both sides of the dividing wall column are operating at
the same pressure, the dividing wall can be a thin wall design (not welded
at the column wall), thus eliminating thermal stress concerns within the
dividing wall column.

Disadvantages of the Dividing Wall Column System

Both dividing wall column systems described above result in taller

distillation columns. This may be perceived as a disadvantage in some units
where site-specific height restrictions apply. However, in general, the
overall height of these columns is still well within normal construction
limits. The combined rectifier/afterwasher, for example, will be about 65
meters tall. This height can be accommodated in most plant locations.

The other disadvantages are associated with the dividing wall column
application in the distillation section. These are higher pressure drop and
higher temperature difference from top to bottom resulting from a single
operating pressure in the column. The operating pressure is set by the
temperature of the condensing medium which is usually cooling water.
Therefore, the top pressure for the dividing wall column is about the same
as the top pressure for each of the conventional columns. Since the dividing
wall column has more trays than each of the separate columns, the
pressure drop is higher than either of the conventional columns.

This higher pressure drop results in a slightly higher temperature in the

bottom of the dividing wall column. The higher temperature requires a
slight reengineering of the heat recovery system; however it can easily be
accommodated and the increase of a few degrees is not expected to have an
impact on fouling in the reboiler.

Designing a Dividing Wall Column System

Given the advantages noted above, why have dividing wall columns not
been used extensively in the past? One major reason is the complexity of
optimizing the design given the number of variables in one distillation
column system. Figure 9 shows the six different sections of a dividing wall
column. Figure 10 lists the parameters that must be considered (11 in total).
Given the number of parameters to calculate and optimize, relatively
sophisticated tools are required to simulate the system. BASF has developed
the proper tools for simulation and optimization of the dividing wall
column system. BASF has also done the required pilot plant work and
laboratory work to verify the simulation model.
The combination of BASF’s experience with dividing wall column
technology in various applications and the advanced simulation tools
allows BASF to optimize the design of dividing wall column systems. The
optimized design ensures proper operation of the column while minimizing
the disadvantages noted above. The wide range of applications in dividing
wall column technology include:

Size:                                 0.6 to 4 m diameter (divided wall columns)

                                          0.7 to 6 m diameter (thermally coupled columns)

Pressure:                      0.002 to 10.0 bar

Purity demands: Simple specifications (e.g. solvent recovery) up to

specifications in the ppb range.

Conclusion
With the ability to optimize dividing wall column designs, BASF has
developed the new butadiene process with dividing wall column
technology that saves on energy consumption and investment cost. This
makes the best technology for butadiene extraction available today even
better.

June 1st, 2012 in Distillation Technology

DESIGN REVIEW AND AUDIT OF ENERGY AND
MATERIAL BALANCE
http://seperationtechnology.com/design-review-and-audit-of-energy-and-material-balance/

Before proceeding with a detailed energy analysis on your distillation unit

as presently operated, you should find out the energy consumption of the
same type of separation by the industry. Your sources of information are:
(1) similar distillation columns within the company, (2) contact with the
original engineering design company, (3) contact with technical people
from your professional groups or college or professional friends, and (4) the
technical literature. For example, Mix, Dweck, and Weinberg estimated and
reported specific consumptions in Btu/lb of product for various product
separations in the CEP April, 1978 issue (see Appendix 7-C). They believe
that a large percentage of the columns in operation can be retrofitted for
energy conservation with attractive economic benefits.

REVIEW OF PLANT DESIGN:

Your plant engineering files should contain all the design information for
the process. If it is not available, this information should be requested from
the original design company. In particular, process flow sheets, design
calculations, piping and instrumentation drawings, specifications of the
equipment purchased, performance characteristics of the equipment,
utility usage tabulations, and revisions since the original installation are
very valuable for the analysis. Examples of process flow sheets are found in
Figures 2-1 and 2-2.

Design values for fuel, steam, and electrical usage should be found on the
utility summary forms. Calculated values for specific operating conditions
should be in the process calculations. Values for fuel and steam usage
should be indicated on the process flow sheet. For example, if the design
values showed 30,000 lbs per hr. of 75 psig saturated steam to produce 6000
lbs per hr of product, the ratio of the pounds of 75 psig saturated steam to
pounds of product is 5. If the condensate is not recovered, the energy usage
is (1185 – 48)5 or 5685 Btu per lb. If a competitor operated with the same
ratio of steam to product, but recovered the condensate at 200º F, his
energy usage is (1185 – 188)5 = 4985 Btu per lb. This is an energy saving of
12%.

Specifications of purchased equipment and their performance are valuable

for any plant study. They must be used with caution because revisions may
have been made since the original installation. If the changes were not
documented (not uncommon in small plants) or simply given verbally to
the present unit supervisor, you may not know that revisions occurred.

AUDIT OF ACTUAL PLANT OPERATION:

After the background information is compiled and the energy information

extracted, the present energy usage of the unit should be determined. Plant
accounting records should be checked for present and past usage of steam,
fuel, electricity, etc. This information may be reported on a monthly basis
on “value added” sheets or “production cost” sheets. All values reported by
accounting should be considered questionable until they can be verified for
accuracy. Instruments may be broken. Flow meters may measure usage for
more than one unit, and the flow split guesstimated. If the guess was
wrong, the estimated values recorded by accounting are in error and could
incorrectly bias your decision on a proposed energy conservation project.

Plant inspections should be made of the measuring instruments. An orifice

meter may have been calibrated for 100 psig line pressure, but the actual
gas pressure found in the plant is 150. The meter’s conversion factor and
reported usage will be incorrect.

Production rates reported by accounting should be confirmed. Production

figures are based upon meter readings and/or product shipments plus
storage tank content changes. A level indicator on a storage tank may be
based upon a 0.800 gravity liquid, but the actual gravity is 0.750. The
production figure is not correct.

A heat and material balance can be made of the existing operation after the
plant instrumentation has been corrected. This information will be
compared with the original design balance and other energy figures found.

DATA COLLECTION DURING PLANT OPERATION:

When developing a heat and material balance for the existing operation,
you may have insufficient information recorded on daily operating and
laboratory logs to compile the balance. Since distillation units are generally
well instrumented, the only expense burdens for a plant data collection test
are the manpower to collect the data and laboratory charges to perform the
analyses on the special samples. Of course, if one flow meter measures
steam usage to two different units, an additional meter must be added to
separate the units.

The degree of success of a plant data collection test is influenced by the

preparation and planning stages. Step one is to list the data required for
calculating the heat and material balance. Measuring locations are marked
on the engineering flow diagram. Step two requires a tour of the unit,
confirming and having calibration checks made of critical measuring
instruments. Dial thermometers, pressure gauges, and dp cells are
examples of these instruments. Following table is an example of a data
collection sheet for electric motors in the unit. When reading pressure
drops across an exchanger, it is preferable to use the same pressure gauge
to read up stream and downstream pressures. A three way selector valve
such as made by D/A Manufacturing Co.,Tulia,Texasis a very convenient
option for making two readings with the same pressure gauge. A more
expensive option is to use a pressure differential transmitter.
The accuracy of flow meters can be checked by the use of a prover, if the
necessary piping manifold is in place or installed. Otherwise, the meter
design calculations and test results made by the instrument department
should be studied and checked. If an orifice meter is in use, you can visually
confirm that the upstream side of the orifice plate is inserted in the line
correctly and that the orifice size stamped on the plate agrees with
specifications. The condition of the orifice opening cannot be checked
unless it is removed.

After all instruments are checked, you can take one data set of readings,
noting time to make readings, and problems in collecting readings or
samples. A heat and material balance can be calculated and inconsistencies
noted. For example, in making an energy balance across an exchanger, the
heat transferred to the colder stream is found higher than the cold stream.
An incorrect temperature reading or flow rate may be the reason. When
this “dry run” is completed and changes made, the plant test and evaluation
are performed.

A  data  collection  run  for  the  electrical  usage  is  determined  by  reading
amperage loads on each motor and reading the wattmeter for the unit over
the test period.  Electric  motors  connected  to  instrument  air  and  plant 
air  compressors should be included in the energy audit.
May 1st, 2012 in Operations of Distillation System
LOW TEMPERATURE SEPARATION
http://seperationtechnology.com/low-temperature-separation/

Over 25 Years Experience

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

LOW TEMPERATURE SEPARATION (L.T.S)

The concept of Low Heat range Separating is based on the ‘Joules-Thomson
Effect’. This is, that when a questionable fluid is extended into a reduced
stress, bigger quantity program, ‘Auto-refrigeration’ occurs resulting in a
loss of temperature.

The level of temperature fall relies on the quantity of stress fall and the
improve in quantity. This concept is used in LTS to get the elimination of
regular water from Organic Gas.

The following simple flower information protects a procedure used on

Organic Gas that is created straight from questionable regular water wells
and therefore needs no pressure before handling.

This procedure information protects only one LTS device. Based on the
variety of regular water wells and quantity of gas to be prepared, the
overall flower will involve several models (or trains).

INLET MANIFOLDS
(Figure: 18)
Each generating well is piped to the inlt shut-down numerous where it goes
through a shut-down device (SDV) which is managed by the automated
urgent shut-down program. Each SDV has a start-up avoid and crammed.

From the shut-down device, the well flow-lines complete to the inlt
numerous where each range is fixed with a circulation factor to gauge the
gas circulation rate. The gas then goes to the flower via a crammed device.

At the inlt numerous the regular water wells can be redirected to the ‘Flow
Splitter’ boat V.1 as managing circumstances require.

FLOW SPLITTER (V.1)

(Figure: 19)

This boat is rounded and provides to individual well liquids from the gas
flow. The main objective of the splitter is to redirect all or part of a well
flow fluid circulation from its regular practice to one or more teaches, by
managing a sequence of valves at the inlt numerous as development needs.

Usually, a LTS program consists of a variety of similar handling teaches.

Each practice is fed by one or two regular water wells to give the preferred
throughput of wet gas.

The splitter program may be managed at, for example, 1700psi.

Figure: 18

The above plan is displaying only one inlt flowline, the other basins are
similar and all may supply the circulation splitter or go straight to one of
the LTS teaches.
Figure: 19

The stress and temperature figures and devices recognition figures used in
this description are simply illustrations and not actually real procedure
information.

The fluid circulation from the inlt numerous and/or the circulation splitter
goes to a well-stream chilly at about 1700 Psi and up to 240 °F. In the
flowline to the chilly, a shut-down device – SDV – is set up which ends on
initial of the shut-down program.

The well-stream chilly is an Air-fin type having two lovers. The store
temperature of the chilly is instantly managed by a Heat range Operator
(TIC) which adapts the message of the fan knives to sustain an store of
about 130 °F. During freezing, cold climate, it may be necessary to closed
down one fan to sustain the needed temperature.

A great store temperature will usually reduce the dew-point depressive

disorders in the LTS models, while a low temperature will cause early
development of hydrates.

The fluid circulation from the chilly is piped to the Advanced Separator (KO
Drum V.2). (Fig: 20).

This is a rounded 3-phase separator boat with a base fluid start. However,
hydrocarbon condensate is re-entrained in the store gas by a ‘Stinger’
procedure which ‘sucks’ the gas condensate out with the gas.

The regular water stage is piped away from the start to the strain headlines
managed by a Stage Operator (LIC) which functions a Stage Management
Valve (LICV) in the strain range. The regular water which is thrown out to
the strain headlines is metered and documented.

Since it is likely that fluid ‘slugs’ will get into the flower from the regular
water wells, V.2 will provide to capture these slugs and reduce the problem
of procedure problems in downstream devices.

The wet gas and condensate keep the top of the boat and complete to the
pipe part of a gas/gas exchanger (# 1) at about 130 °F and is chilled by
freezing dry gas from the LTS to about 95 °F. In the range to the exchanger,
a SDV is set up to release liquids to width on initial of the shut-down
program.

The tube-side wet gas store temperature of the exchanger is managed by a

Heat range Operator managing a device. This is a 3-way device including a
by-pass, placed in the freezing gas range from the LTS to the shell-side of
the chilly.

A differential stress signal evaluate – PDI – actions and indicates the

Pressure Distinction (PD) across the chilly and is used for regular
assessments by managing employees.

Normal PD should be up to 15 psi and, if it starts to improve, may indicate

the development of hydrates or wax in the pipes.

The chilled, wet gas goes into the Choke Knock-out boat V.3.

July 1st, 2012 in Crystallization Process

New Technology in Rich Gas Absorption
http://seperationtechnology.com/new-technology-in-rich-gas-absorption/
Coursary
oursary
Search for

courses for
Coursa
free
Open

The McCabe-Thiele method for analysing binary absorption and stripping

problems is taught as a standard part of most undergraduate chemical
engineering degrees. The case of mutually insoluble carrier streams is
usually considered first, with the problem set out as illustrated in Figure 1.
V and L are the mole (or mass) flow rates of the vapour and liquid streams
(or light and dense streams in the case of liquid-liquid extraction), a and b
refer the top and bottom of the column, and x and y are the mole (or mass)
fractions of the soluble component in streams L and V respectively [1].

The operating line is obtained by performing a material balance around

stages 1 to n of the column:
In general this is a curved line, although for lean-phase systems (where the
concentration of the soluble component is always less than 10%) the
magnitude of the stream flowrates, L and V, remain nearly constant and so
the operating line can be approximated as a straight line:

Students are often required to determine the theoretical minimum liquid

flowrate, La,min, required in order to achieve a desired separation. This is
the liquid flowrate at which an infinite number of stages would be required
and it occurs when the operating line just touches the equilibrium line (the
“pinch” point).

For lean-phase problems, the solution is trivial. The student only needs to
draw a straight operating line from the known conditions at the top of the
column (xa,ya) which just touches the equilibrium line. If the equilibrium
line is also straight (eg. Henry’s law y= mx) then this will occur at end b of
the column where Lb leaves in equilibrium with the entering gas Vb at the
point xb=yb/m (Figure 2a).

However, for rich phase systems (concentrations greater than 10%) the
operating line has significant curvature because the ratio of L to V varies
down the length of the column. In this case, assuming the required
operating line passes through the point (yb/m, yb) may not always be
correct. If the operating line is concave up then it is possible that the
operating line may cut the equilibrium line at some point between ya and
yb (Figure 2c), and so the operating line through (yb/m, yb) gives too low an
estimate of La,min. What is more, the student only becomes aware of this
error if they take the time to plot the operating line for the minimum liquid
flowrate case, whereas they often without checking go on to solve the main
part of the question which involves calculating the number of stages
required when La is some multiple of La,min.

In this case where the operating line is concave up (or concave down for
stripping problems), to find the operating line which just touches the
equilibrium line the students must either adopt a lengthy trial and error
approach, or else they must solve the problem graphically [2,3] by first
converting the problem into mole (or mass) ratios, X and Y, where X = x/(1-
x) is the moles of solute per mole inert carrier fluid and Y = y/(1-y) is the
moles of solute per mole of inert carrier vapour (0 < X,Y < ¥). When mole
ratios are used, the liquid and vapour flow rates are given as =(1-xa)La and
=(1-y)Va, the moles of solute-free liquid and vapour flow respectively. For
mutually insoluble solute streams,  and  remain constant, and so the
operating line is a straight line given by the equation:

Once the equilibrium data has also been converted into mole ratios and
plotted, the minimum condition can easily be found graphically using a
ruler to find the straight line starting at (Xa, Ya­ ) which touches the
equilibrium line. The slope of this line is , from which La,min can be
calculated.

This paper presents a new analytical approach for finding the minimum
liquid flowrate in rich phase problems which does not require converting
the problem into mole ratios. The new method requires that an analytical
expression for the equilibrium line be known and that this remains
constant through the length of the column ie. the column must be operating
isothermally. This method involves re-arranging equation (1) into an
explicit expression for y in terms of x and solving to find the point at which
it just touches the equilibrium line. This analytical method can be taught to
students to complement the traditional graphical approach.

The New Method

In most text books, the equation for the operating line is left as shown in
equation (1). This equation cannot be used immediately to solve for y in
terms of x, because L and V are both also functions of x for the rich phase
case.

Since the flow of inert carrier fluid remains constant for mutually insoluble
streams, (1-x)L = (1-xa)La. Therefore:

A total material balance around stages 1 to n gives V = L + Va – La, which

using equation (4) may be re-written as:
Substituting equations (4) and (5) into equation (1) and re-arranging gives
an explicit equation for y in terms of x as the only variable:

Equation (6) can also be differentiated to give the equation for the slope of
the operating line at any point:

Equation (6) is simple to derive, requiring only algebraic substitution and

re-arrangement of equation (1) or (3). However, although trivial to derive, it
is not presented in this form in any of the standard introductory texts on
separation processes [1-5]. Its usefulness lies in the fact that as an explicit
function for y in terms of x, it is easy to differentiate, giving equation (7)
which is novel. Equations (6) and (7) are extremely useful because they can
be used directly to solve for y and dy/dx at any point down the column in
terms of only one variable, x. Choosing end a of the column as the reference
point was arbitrary. These equations can equally well be written in terms of
end b (or in terms of any other known point along the length of the column)
by simple substitution of Lb for La, xb for xa, etc.

Provided that we have an analytical expression for the equilibrium

relationship which is constant through the length of the column, these two
equations can now be used to analytically find the minimum liquid
flowrate, La, min, required to achieve a given separation. The simplest case
where the equilibrium line is given by Henry’s law (y* = mx),  will now be
considered as an example.

La,min occurs when the operating line and equilibrium line touch at a single
point between ya and yb. This intersection point can be found analytically
and is given by  (see Appendix A):

where a = (1-mxa)2, b = 4m(ya+xa)-2(m+ya)(mxa+1) and d = (m–ya)2. For

stripping problems where the operating line is below the equilibrium line,
Vb,min is found by equation (9):
where a = (1-mxb)2, b = 4m(yb+xb)-2(m+yb)(mxb+1) and d = (m–yb)2.

To decide whether using equation (8) or (9) is necessary, it is first required

to check whether or not the pinch point lies between ya and yb, or at yb. The
general solution strategy for finding La,min in any rich-phase gas absorption
problems is as follows.

Step 1. Begin by assuming that the pinch point where the operating line just
touches the equilibrium line is at the base of the column. Therefore xb =
yb/m and La is found by an overall material balance around the column:

In                     = Out              (at Steady State).

Solute:              ybVb + xaLa       = xbLb + yaVa

Inert Liquid:      (1-xa)La            = (1-xb)Lb

Re-arranging and solving for La gives:

 Step 2. Calculate the slope of the operating line at xb = yb/m using equation
(7).

Step 3a. If the slope of the operating line at xb = yb/m is less than the slope of
the equilibrium line (i.e. dy/dx <  m). This indicates that the operating line
has crossed the equilibrium line from above, as shown in Figure 2b, so the
La found in Step 2 is the correct La,min.

Step 3b. If the slope of the operating line is greater than the slope of the
equilibrium line (i.e. dy/dx >  m), then this indicates that the operating line
is intersecting the equilibrium line from below, as shown in Figure 2c. In
this case equation (8) is then be used to find the correct minimum liquid
flowrate.

For stripping problems, the operating line lies below the equilibrium line
and the full conditions are known at end b, but not end a. The aim is to find
Vb,min and the requirements for the slope of the operating line at the point
of intersection are reversed.  The solution procedures for both absorption
and stripping problems are summarised in Table 1. A worked example
problem illustrating both this solution procedure and the traditional
approach is given in Appendix B.

 
The above solution procedure can be easily adjusted to consider other
analytical expressions for the equilibrium line, y* = f(x). If the equilibrium
line is given by equilibrium data which does not readily fit any simple
analytical expression, then the student has no choice but to convert the
problem into mole ratios and solve graphically.

Discussion
The new method proposed is fully analytical. However, the intuitive
understanding behind the derivation which students need to appreciate is
based on a graphical understanding of the problem. Hence, it cannot
replace the traditional graphical approach using solute-free coordinates. It
is, however, complimentary and provides students with a different set of
tools for tackling such problems. In addition, the derivation of this method
serves to remind students that the basic tools of analytical geometry they
learnt at school can be applied to apparently unrelated engineering
problems.

Although no formal survey of students was performed, the author’s

informal impression after lecturing one class has been that many of them
preferred to use analytical expressions, rather than having to convert the
problems into mole ratio units and then use graphical methods.

Equation (6) is not only useful for finding the point at which the operating
line touches the equilibrium line. It can also be used to help plot the curved
operating lines that occur in any rich-phase problem. This is required in
order to be able to step off the number of stages via the McCabe-Thiele
method, or to perform the numerical integration required to find the
number of transfer units in a packed column.

Even if the full analytical method is not used, the equation for the slope of
the operating line, equation (7), is valuable because it enables students to
test whether the end point (yb, yb/m) is the correct pinch point without
having to plot the full operating line.

Equations (8) and (9) are also potentially useful for software packages for
computer based learning packages where each student in a class can be
given different computer generated problems to solve independently and
then enter their answers into the computer for checking.

Conclusion
Explicit equations for the operating line and its slope in rich-phase gas
absorption and stripping problems have been derived with x as the only
variable. These expressions, although trivial to derive, are not presented in
any of the standard introductory texts on separation processes. They have
been used to develop a new analytical method for finding the minimum
liquid flowrate in rich-phase problems without needing to convert the
problem into solute-free coordinates and then use graphical methods.

May 2nd, 2012 in Adsorption

PRODUCING HIGH PURITY BTX (Benzene-Toluene-
Xylene)
http://seperationtechnology.com/producing-high-purity-btx-benzenetoluenexylene/

Trillium Flow Technologies

Application-Specific Pumps
Complete Project Support Through Each

Stage Of The Life Of Your Operation.

Contact Us Now.

trilliumflow.com

OPEN

Producing High purity BTX components (benzene, toluene and xylenes) are
important building blocks in many chemical processes. In particular,
considerable quantities of benzene and paraxylene are used in the
production of major petrochemical intermediates for the polymer industry.

Naphtha cracking produces light olefins for polymer applications, but also
produces pyrolysis gasoline (pygas), which is rich in aromatic compounds
and a significant source of BTX. Pygas has been a component in the gasoline
pool, but is increasingly used for its higher petrochemical value. This
requires the separation of BTX components, via extraction, from aliphatic
compounds and C9+ aromatics for producing high purity BTX.

In many cases, naphtha crackers have been extensively debottlenecked to

add capacity; however, existing extraction units do not always have the
flexibility to support significant capacity

increases. Also, due to aromatics limitations in the gasoline pool, the high
value of ethylene, propylene and high purity BTX intermediates, the idea of
converting C7+ pygas to these highly desired components is an attractive
option.

THE SK – AXENS APU COMBINATION

SK and Axens have teamed up to respond to market demand by providing
the product that will do exactly this. APU (Advanced Pygas Upgrading) is a
catalytic process technology developed by SK Corporation and exclusively
offered by Axens to convert pygas to a superior olefin feed (LPG) and BTX
aromatics.

Cuts originating from 2nd Stage Pygas Hydrogenation units are employed
as feedstocks. The principal catalytic reactions are:
o Conversion of non-aromatics (especially C6 to C8
alkanes) into ethane and
LPG.

o Conversion of C9+ aromatics into BTX, thereby increasing BTX yield.

The high activity and selectivity of the catalyst used in this application,
APU-11, provide a number of significant benefits:
o Production of a pure benzene cut

o Production of a pure toluene cut

o Production of a xylenes cut with

low ethylbenzene content
o Enhanced BTX yield through C9+ cut conversion

o Co-production of LPG
o Potential to increase ethylene and propylene by recycling ethane and LPG
to the steam cracker furnaces

As compared to pygas extraction followed by hydrodealkylation (HDA), the

APU exhibits a differentiated product pattern as illustrated graphically
below.

feedstock, and the xylenes fraction does not require further extraction
before separation to recover paraxylene.

In some locations, ethane and LPG are the desired products; in others they
provide cracking furnace feedstocks. Typical olefin yields based on the
original pygas feed are 12.5% are 3.2% for C2= and C3=, respectively.

The APU process employs moderate operating conditions and displays

competitive CAPEX & OPEX as compared to extraction and HDA. The
process is well suited for the reutilization of existing assets in refineries.
APU technology is the ideal choice for:

o Complementing/debottlenecking existing Extraction Units for the

production of high purity aromatics

o Converting low value pygas, especially the C9+ fraction often sent to fuel
oil, into BTX, C2 and LPG
o Increasing C2= and C3=
production by recycling C2-C4 paraffins to the
cracking furnaces The following slides and illustrations present a more
detailed review of the process and economic advantages of APU.

May 21st, 2012 in Design Distillation System

C2 Splitter Aspen Model for Troubleshooting
http://seperationtechnology.com/c2-splitter/

Over 25 Years Experience

Wet filters, 3N1 coalescing filters, salt water removal,

& more. Call now.

Sepra-Chem Corporation Open

C2 Splitter is usually the final step in ethylene manufacturing process. The

article below describes a hypothetical case study of C2 fractionation unit
using Aspen Plus as Process Simulation tool for troubleshooting operations
of C2 Splitter.

Introduction

The Olefinas capacity of 250 000 TPA of Ethylene from feed ethane and
propane. To achieve this capacity, a two column C2 Splitter 110El/110E2 was
installed, to handle 37% of plant load.

The old unit 107E was retrayed to handle 63% of the plant load. Column
overhead gas at 16.8 barg is fed to a heat pump compressor, and retumed at
higher pressure of 35 barg to supply heat for reboiling. A sketch of this
system appears below ( Fig. 1 ).
The problem of high pressure drop across C2 Splitter 107E:

The Olefinas I Plant has built in flexibility for handling ethane and propane
feeds, but historically had problems in operating at design load.

The expected loads to 107E for the three design cases are tabulated in Table
1.
Historically, at high load there were frequent increases in pressure
differential across the column. Often, this was related to hydrate formation,
which was handled by decreasing the load and injecting Methanol into the
column. Dehydrator regeneration was improved; but incidents of
increasing pressure drop in 107E continued.

AspenPlus was used for simulating this column for better understanding
the system. AspenPlus predicted a Flood Factor of 0.89 for the Typical Feed
Case and 0.99 at the highest design load, corresponding to the Mixed Feed
Case. This raised doubts of the adequacy of design of this system. The
process licensor was contacted to obtain an explanation. When it was
confirmed that the design was adequate, a systematic check was carried out
to establish the cause for the high pressure drop.

Causes of increased pressure drop across 107E; and corrective action:

1. Hydrate formation in the column, due to moisture break through:

Initially, this was due to inadequate regeneration of the molecular sieves in

the dehydrator, or due to problems during switching of dehydrator beds.
This was corrected. Residual moisture retained in the cold box would also
get released whenever there was a "Warming" of the cold box. These
incidents continued for some time, and petered out.

2. Damagteo trays, or partial blockage:

Since the high pressure drop had been experienced for many years, a
partial blockage of some trays was suspected. It was decided to get the
column scanned by TRU-TEC. This involved scanning the column
throughout its length during stable operation, using a technique that has
been very effectively applied by TRU-TECA. radioactive source and a
detector are positioned outside the column, across the active area of the
trays in a column, and lowered simultaneously, in small steps. The detector
signal varies, depending on what is between the source and the detector.
This is recorded in a portable computer, and later printed out and
interpreted.

A scan at low load was carried out to confirm that all the trays were in
place. The reflux was increased to flow-rates that were often problematic,
and another scan carried out which confirmed that there was no alarming
abnormality. Simulation of the column during this scan provided a valuable
insight on how to use AspenPlus effectively.

A scan at high load was planned and carried out, just prior to a plant
turnaround. This achieved stable operation for a short period of time,
which confirmed that there was no abnormality in the column. It was
recommended that the trays in the bottom section of the column be checked
out during the plant turnaround.

Improved instructions for column operation:

Armed with the valuable insights gained during the scan, and with a
validated simulation, the operation of this system was critically reviewed.

1. Correction of inadequate reboiling:

The system was being operated at ‘ design conditions ‘, which included

maintaining a backpressure of 32 barg downstream of the reboiler
condensate pot. At high load operation,this resulted in restricted removal of
condensed ethylene from the reboiler pot even with the level control valve
fully open, which lead to inadequate reboiling. Column load was therefore
restricted, or whenever load was increased it was accepted that there
would be some loss of ethylene to the column bottom stream.
A backpressure of 30 barg was suggested, which helped removal of
condensed ethylene even at the highest load, and ethylene loss due to
inadequate reboiling was eliminated.

2. Choice of feed tray:

For two defined design cases, the feed was to be introduced at tray 64 or at
tray 74. Since there were frequent variations in feed availability to the
plant, and since actual operation was inbetween design cases, the feed
location was being maintained at tray 74. It was shown that by maintaining
reflux to feed ratio above 3.0, feed could always be introduced at tray 64.
Loss of ethylene to column bottom stream was reduced, due to this
simplified operation.

Cheeidng the column during plant turnaround:

During the plant turnaround in March 1997, the column 107E was opened
for inspection. Here was some dirt in the bottom section, between trays 81
and 85 which reduced the active area by around 5 percent. These were
cleaned; there was no other abnormality in the column. Whenth e columnw
as again in operation, close to the typical feed case, it was loaded to around
110 percent of the feed estimated for this case. The system operated stably,
and has since operated at higher than design capacity on many occasions.

Using AspenPlus more effectively:

Actual operation of the column was simulated using AspenPlus (see Table 2,
below).
It was seen that the pressure drop predicted by simulation was always less
than the actual pressure drop. Also, the flood factor predicted was always
high; but from the experience of previous evaluations, it was knownth at
this was not a matter of concern.

Gains from this troubleshooting effort:

While this troubleshooting was carried out over a long time span, there
were well defined gains:

1. The problem of high pressure drop across 107E was due to hydrate
formation. With improved operation, these incidents have been all but
eliminated.

2. The plant has a nameplate capacity of 250 000 TPA, or 750 T/day.
Previously, production was limited to 80% of this capacity, because of some
other problems and suspected C2 splitter column design inadequacy. Now,
the same column operates at higher than design expectation, and the plant
has operated on several occasions to produce more than 800 T/day,
reaching 850 T/day on one occasion.
3. Scanning of columns in operation is a well known tool for
troubleshooting. This was effectively used for eliminating suspicions of
fallen or partially blocked trays.

4. Simulation tools have some limitations. During this troubleshooting

effort, taking advantage of scanning of the column, the limitations of
AspenPlus in predicting flood factor were discovered and simulations of
actual operations were successfully validated. Experience has been gained
on the use of AspenPlus to make it a powerful tool for successful
troubleshooting.

The final step in ethylene purification during its commercial production is

usually the separation of ethylene from ethane by fractionation, in a C2
splitter column. This paper outlines the operating problems that were faced
in C2 splitter columns in two different Ethylene Plants,and how
troubleshooting was carried out successfully. Additionally, a valuable
insight was obtained on using AspenPlus effectively for such systems,
which was further consolidated during test runs, makingit a powerfultool
for effective troubleshooting.

Introduction

The Olefinas I Plant in the Zulia Complex of PEQUIVE was revamped in

1989, to a nameplate capacity of 250 000 TPA of Ethylene from feed ethane
and propane. To achieve this capacity, a two column C2 Splitter 110El/110E2
was installed, to handle 37% of plant load.

The old unit 107E was retrayed to handle 63% of the plant load. Column
overhead gas at 16.8 barg is fed to a heat pump compressor, and retumed at
higher pressure of 35 barg to supply heat for reboiling. A sketch of this
system appears below ( Fig. 1 ).

The problem of high pressure drop across C2 Splitter 107E:

The Olefinas I Plant has built in flexibility for handling ethane and propane
feeds, but historically had problems in operating at design load.
The expected loads to 107E for the three design cases are tabulated in Table
1.

Historically, at high load there were frequent increases in pressure

differential across the column. Often, this was related to hydrate formation,
which was handled by decreasing the load and injecting Methanol into the
column. Dehydrator regeneration was improved; but incidents of
increasing pressure drop in 107E continued.

AspenPlus was used for simulating this column for better understanding
the system. AspenPlus predicted a Flood Factor of 0.89 for the Typical Feed
Case and 0.99 at the highest design load, corresponding to the Mixed Feed
Case. This raised doubts of the adequacy of design of this system. The
process licensor was contacted to obtain an explanation. Whenit was
confirmed that the design was adequate, a systematic check was carried out
to establish the cause for the high pressure drop.

Causes of increased pressure drop across 107E; and corrective action:

1. Hydrate formation in the column, due to moisture break through:

Initially, this was due to inadequate regeneration of the molecular sieves in

the dehydrator, or due to problems during switching of dehydrator beds.
This was corrected. Residual moisture retained in the cold box woulda lso
get released whenevert here was a ?Warming of the cold box. These
incidents continued for some time, and petered out.

2. Damagteo trays, or partial blockage:

Since the high pressure drop had been experienced for many years, a
partial blockage of some trays was suspected. It was decided to get the
column scanned by TRU-TEC. This involved scanning the column
throughout its length during stable operation, using a technique that has
been very effectively applied by TRU-TECA. radioactive source and a
detector are positioned outside the column, across the active area of the
trays in a column, and lowered simultaneously, in small steps. The detector
signal varies, depending on what is between the source and the detector.
This is recorded in a portable computer, and later printed out and
interpreted.

A scan at low load was carried out to confirm that all the trays were in
place. The reflux was increased to flow-rates that were often problematic,
and another scan carried out which confirmed that there was no alarming
abnormality. Simulation of the column during this scan provided a valuable
insight on how to use AspenPlus effectively.

A scan at high load was planned and carried out, just prior to a plant
turnaround. This achieved stable operation for a short period of time,
which confirmed that there was no abnormality in the column. It was
recommended that the trays in the bottom section of the column be checked
out during the plant tumaround.

Improved instructions for column operation:

Armed with the valuable insights gained during the scan, and with a
validated simulation, the operation of this system was critically reviewed.

1. Correction of inadequate reboiling:

The system was being operated at ‘ design conditions ‘, which included

maintaining a backpressure of 32 barg downstream of the reboiler
condensate pot. At high load operation,this resulted in restricted removal of
condensed ethylene from the reboiler pot even with the level control valve
fully open, which lead to inadequate reboiling. Column load was therefore
restricted, or whenever load was increased it was accepted that there
would be some loss of ethylene to the column bottom stream.

A backpressure of 30 barg was suggested, which helped removal of

condensed ethylene even at the highest load, and ethylene loss due to
inadequate reboiling was eliminated.

2. Choice of feed tray:

For two defined design cases, the feed was to be introduced at tray 64 or at
tray 74. Since there were frequent variations in feed availability to the
plant, and since actual operation was inbetween design cases, the feed
location was being maintained at tray 74. It was shown that by maintaining
reflux to feed ratio above 3.0, feed could always be introduced at tray 64.
Loss of ethylene to column bottom stream was reduced, due to this
simplified operation.

Cheeidng the column during plant turnaround:

During the plant turnaround in March 1997, the column 107E was opened
for inspection. Here was some dirt in the bottom section, between trays 81
and 85 which reduced the active area by around 5 percent. These were
cleaned; there was no other abnormality in the column. Whenth e columnw
as again in operation, close to the typical feed case, it was loaded to around
110 percent of the feed estimated for this case. The system operated stably,
and has since operated at higher than design capacity on many occasions.

Using AspenPlus more effectively:

Actual operation of the column was simulated using AspenPlus (see Table 2,
below).

It was seen that the pressure drop predicted by simulation was always less
than the actual pressure drop. Also, the flood factor predicted was always
high; but from the experience of previous evaluations, it was knownth at
this was not a matter of concern.

Gains from this troubleshooting effort:

While this troubleshooting was carried out over a long time span, there
were well defined gains:

1. The problem of high pressure drop across 107E was due to hydrate
formation. With improved operation, these incidents have been all but
eliminated.

2. The plant has a nameplate capacity of 250 000 TPA, or 750 T/day.
Previously, production was limited to 80% of this capacity, because of some
other problems and suspected C2 splitter column design inadequacy. Now,
the same column operates at higher than design expectation, and the plant
has operated on several occasions to produce more than 800 T/day,
reaching 850 T/day on one occasion.

3. Scanning of columns in operation is a well known tool for

troubleshooting. This was effectively used for eliminating suspicions of
fallen or partially blocked trays.

4. Simulation tools have some limitations. During this troubleshooting

effort, taking advantage of scanning of the column, the limitations of
AspenPlus in predicting flood factor were discovered and simulations of
actual operations were successfully validated. Experience has been gained
on the use of AspenPlus to make it a powerful tool for successful
troubleshooting.

May 15th, 2012 in Design Distillation System, Operations of Distillation

System, Process Simulation
Graphite Heat Exchanger
An ISO 9000 certified company, we

pride in maintaining best-in-class

standards and quality

Split Tower –
The feed stream will be split so that the condenser duties of the high
pressure column approximately matches the required reboiler duty of the
low pressure column. (See Figure 4-3 for an example split tower
arrangement). The heat, input to the reboiler, of the high pressure column
rises to the condenser where it then provides the reboiling duty of the other
column. By use of the split tower arrangement, we have cut our energy use
almost in half. Note that instead of two columns, any number of columns
can be used in the split tower fashion. However, For each additional tower,
an extra delta T must be supplied, plus the temperature drop across the
column. In addition, the energy savings drops as each column is added. The
two tower system saves 50% of the energy. Another tower saves (50 – 33) or
only 17%. A fourth tower will save only 8.3%. So our write up will deal with
only the two column arrangement.

The split tower system has a single reboiler and single condenser. The
temperature difference between the reboiler and condenser will be much
greater than that of an ordinary column. This occurs because the two
columns each have their own temperature difference to be met from the
top to bottom, and the driving force for the condenser reboiler must be
supplied. As a result of this for the split tower arrangement to work, the
following factors must be present:

1. The temperature and pressure in the high pressure column must be

below the critical points.

2. The pressure must not be so large as to require too heavy column walls.
3. The low pressure column must not be too low, so low a vacuum as to
cause trouble.
4. The products must not be degraded by the highest temperature or
frozen, or too viscous at the lowest temperature.
5. The heat source must be able to supply heat at a temperature above that
of the reboiler.

6. The condenser temperature must not be below that obtainable by

conventional air and water cooling. Refrigeration cannot be tolerated,
unless the conventional column would also need refrigeration.
The split tower arrangement has a large temperature difference between
the reboiler and condenser, thus it will probably be desirable to minimize
this by using small delta T’s across the reboiler, condenser-reboiler, and
condenser. This will mean a large heat exchanger surface being required.
Even so, it is likely a higher temperature heat source will be needed for the
reboiler. As it is at a higher temperature, the heat will be more expensive,
such as a higher pressure steam. This means we are saving energy, but
using a more costly source.

The feed to a single tower will be split in two for the two column
arrangement. Therefore, the individual columns will be about one-half the
size of the single column. However, the relative volatility and the mass
flowrate/area through the columns will change with the pressure, resulting
in a differently sized tower than just one-half the size.

From an economical viewpoint a split tower arrangement will require two

columns, instead of one larger one. Each column will require its own
instrumentation, causing twice the instrument costs. The higher pressure
column will need thicker walls, and its size may be larger than expected.
(See preceding paragraph). A larger exchange surface is needed for the
various exchangers. Various auxiliary exchangers may be required for
column control. The savings of the split tower arrangement come from the
reduced heat requirement. However, the value of the heat used should be
higher per Btu used than in the case of a single column. In many respects, a
split tower will be similar in economic desirability to a vapor re-
compression column. The key is to have a low temperature difference from
the top to the bottom of the column.

In designing the split tower arrangement, the low pressure column should
be set by the achievable condenser temperature. Then the split tower
should be worked backwards from this point, a reasonable temperature
drive given for the condenser-reboiler, then the high pressure column
found, finally resulting in a temperature for the reboiler. With this
temperature the available heat sources should be examined, for example,
the various steam pressures, and one chosen. The delta T available should
then be distributed between the reboiler, condenser-reboiler, and
condenser to obtain the minimum required heat exchanger surface area.
The feed between the towers should be split in order to approximately give
equal duties for the high pressure condensation and low pressure reboiling
under design operating conditions.

The control of a split column will be more complex than that of a single
column. The object of the control system will be to decouple the two towers
to a certain extent. The use of an auxiliary condenser on the high pressure
column and a auxiliary reboiler on the low pressure column will give
energy efficiency and good control. Control can also be had by having only
one auxiliary exchanger, and by having one of the columns run at a higher
duty than the other. The feed split ratio between the columns can be used
as part of the control. Note that the bottoms of the high pressure tower can
be mixed with the low pressure bottoms and flashed in the low pressure
tower. This would result in a uniform bottoms composition.

The split tower design offers a good possibility of energy savings with a
new installation. Where an existing column exists already, it would be
possible to increase capacity by adding another tower next to the existing
one and installing a new condenser-reboiler so that the existing column
will become one-half of a split tower arrangement. In cases where no
capacity increase is desired, but the column original size was such that two
towers were used, it may be possible to convert it to a split tower operation
by installing a more efficient column internal trays and by adding a
condenser reboiler, new piping, and new instrumentation. The savings that
can result from a split tower design are very much affected by the cost of
energy to the reboiler, so the true energy cost should be evaluated before
using a split tower.

February 6th, 2012 in Design Distillation System

VAPOR RECOMPRESSION AND HEAT PUMPS FOR
DISTILLATION
http://seperationtechnology.com/vapor-recompression-and-heat-pumps-for-distillation/

Trillium Flow Technologies

Highly Engineered Pumps
Precision Engineered To Delivery Long-Term

Success As Your Partner In Flow Control.

trilliumflow.com

OPEN

Vapor recompression consists of taking the overhead vapors of a column,

condensing the vapor to liquid, and using the heat liberated by the
condensation to reboil the bottoms liquid from the same column. The
temperature driving force needed to force heat to flow from the cooler
overhead vapors to the hotter bottoms product liquid is set up by either
compressing the overhead vapor so it condenses at a higher temperature,
or lowering the pressure on the reboiler liquid so it boils at a lower
temperature, then compressing the bottoms vapor back to the column
pressure. A conventional column has a separate condenser and reboiler,
each with its own heat transfer fluid such as cooling water and steam. The
vapor recom-pression column has a combined condenser-reboiler, with no
external heat transfer fluids.
The vapor recompression cycle has a set ratio between available condenser
side to reboiler duty. The reboiler heat flow obtained will be equal to the
sum of condenser duty plus the work added to the gas stream and its
inefficiencies. In all cases where the column reboiting and condenser duties
do not match in the manner stated an auxiliary system will be needed to
supply the excess column condenser or reboiler duty.

The advantage of vapor recompression lies in its ability to move large

quantities of heat between the condenser and reboiler of the column with a
small work input. This results from cases where there is only a small
difference between the overhead and bottoms temperature. A conventional
column with steam heating and water cooling may use ten times the Btu’s
of a column running with vapor recompression.

January 21st, 2012 in Design Distillation System

Guidelines for Considering Vapor Recompression
http://seperationtechnology.com/guidelines-for-considering-vapor-recompression/

The following guides indicate when to consider a vapor re-compression

system in comparison with a conventional system for a distillation column.

1. When the column is to operate wholly in refrigeration temperatures for

both the condenser and reboiler, then vapor recompression should
always be considered.
2. Where the condenser temperature is expected or desired to be below
100°F, some consideration of vapor recompression should be given if the
condenser-reboiler temperatures are close, say less than 40°F.
3. Where the reboiler temperature is very high, +500°F, and the
temperature difference is moderate (<100°F).
4. Cases where the condenser and reboiler temperatures are nearly the
same (<20°F) so that high efficiencies can be projected.

5. Where some problem with supplying steam exists for a conventional

column and reasonably good efficiencies can be expected, say 3, 4 to 1
heat/work ratios.

6. Where water is the main fluid, with trace quantities of contaminates

being concentrated so that only a small temperature difference is
expected. Water is an excellent fluid for vapor recompression as it is a
good refrigerant and has a very high heat transfer coefficient.
Two cases for not using vapor recompression are as follows:

1. When larger temperature differences are expected between the

condenser and reboiler.

2. When the column can be operated with cooling water for the condenser
and waste heat (1 ATM or less steam) for the reboiler.

January 24th, 2012 in Design Distillation System

Advantages & Disadvantages of Vapor Recompression
http://seperationtechnology.com/advantages-disadvantages-of-vapor-recompression/

theGB4000.com
Thinking Of The BCX Ultra?
Free Shipping And Free Documentary

Video. Request Your Information Now.

thegb4000.com

OPEN

Advantages of Vapor Re-compression

The advantages of vapor re-compression are:

1. A large amount of heat can be transferred with little work. For example,
suppose we had a heat/work ratio of 10. This means that for every Btu of
work added, we replace 10 Btu’s of heat that would otherwise be
required. This can lead to overall savings also. Assume we generate one
Btu of electricity from three Btu’s of fuel heat (33% efficiency). Then
overall we save 10/3 = 3.33 Btu of heat for every Btu fuel used by the
vapor re-compression system over that of the conventional system.
2.  The heat flow to supply the condenser or reboiler may be of particular
value. For example if the column is cold, the cooling must be
accomplished by expensive refrigeration. This is also true at high
temperatures 500°F+ where energy added to the reboiler can no longer
be supplied simply by steam. The vapor re-compression cycle, by being
balanced, is affected only to a limited effect by the relation of the actual
temperature to the ambient. The pressure in a vapor re-compression
column can be set where desired to achieve maximum separation.
 3. By freeing the condenser and reboiler of the desire to hold temperature
between a minimum of about 100°F to a maximum of about 500°F, the
points easily reached by cooling water and condensing steam, we can
set the temperature, and therefore the pressure, at any point we wish.
This effect is of particular importance where changing the pressure
effects the relative volatility. By operating at more favorable conditions
we can reduce the reflux requirement and therefore the heat duties. The
effect of the pressure change on the column will be to change the wall
thickness, and column diameter, the diameter falling as the reflux is
reduced, but increasing as the pressure drops. Do not neglect the effects
of pressure changes on the overall column.

4. Work energy from excess high pressure steam which is generated for
low pressure steam requirements may be available in the overall plant
balance. This energy would be very cheap if the alternative or current
practice is to wastefully let down the steam across a valve. A steam
turbine driver for the vapor compressor can be used, giving a low work
energy cost.
5. Electricity can be brought in from the outside to run the compressor
driver, so that large amounts of steam are not needed. Thus the steam
plant can be smaller.
6. The vapor re-compression system uses little cooling water or steam flow.
Possible utility savings can occur as less cooling water is discharged, and
the steam, condensate, and cooling water lines can be made smaller.
7. Swings in the ambient temperature and weather will have little effect on
the operation of vapor re-compression.

Disadvantages of Vapor Re-compression

The disadvantages of vapor re-compression are:

1. Premium electrical or steam pressure work energy is used for driving

the compressor, and no advantage can be taken of possible existing low
value sources of waste heat to run the reboiler.
2. The additional cost for the compressor and its driver are required.
3. The condenser-reboiler of vapor re-compression has the overhead and
bottoms product on either side, leading to savings by having one half of
the total area for the separate condenser and reboiler of a conventional
column. This advantage is lost and more heat exchanger surface is
required for the condenser reboiler as compared to the conventional
condenser and reboiler because:

The fluid used (cooling water and steam) in the reboiler and
condenser will have a much higher heat transfer coefficient than the
column fluids.
The EX TD driving force for both the reboiler and condenser will be
much higher than that allowed across the condenser rebolier as its
heat/work efficiency drops with increasing T D across the system.
The EX TD across the condenser-reboiler causes a loss in efficiency,
therefore requiring a larger compressor, driver, and work input. The
EX TD will be minimized to allow a smaller compressor and driver.
These items add up to the fact that the exchanger surface required
for the condenser-reboiler will probably be significantly larger than
the combined surface on the condenser and reboiler in a
conventional column.
4. Auxiliary units such as extra reboilers, condensers will be required in
order to balance the heat duty and control the column.
5. The mechanical complexity of the vapor compressor is high, the system
requiring more maintenance and suffering more breakdowns.
6. More instrumentation will be required to control the compressor and
the other auxiliary items of the vapor re-compression system.
7. The control of the vapor re-compression system is different than that
faced in an ordinary column. New methods will have to be learned by
the operating personnel.
8. Flexibility is lost as the column has only limited ability to function at
other than design conditions. The compressor will be sized to be most
 efficient at one operating rate. Increasing the column reflux over design
will be very hard.
9. Altering the column for reuse to operate for a new situation will be
more difficult than with a conventional column, as the vapor re-
compression system will probably need to be replaced.
10. Continuous auxiliary refrigeration may be required for a low pressure
column with vapor re-compression that would be run at higher
temperature by a conventional distillation system.

Overall, by switching to vapor re-compression one gains energy efficiency

at the cost of greater mechanical complexity and flexibility loss. Capital
costs for vapor re-compression will probably be greater than a
conventional case, but it depends on the situation and what all is taken into
account in economic changes.

January 30th, 2012 in Design Distillation System

Benzene Distillation
http://seperationtechnology.com/benzene-distillation/

Trillium Flow Technologies

Highly Engineered Pumps
Precision Engineered To Delivery Long-Term

Success As Your Partner In Flow Control.

trilliumflow.com

OPEN

Benzene Toluene Fractionation to produce pure benzene via Benzene

Distillation during the past decade, multivariable controllers for have been
extensively applied in ethylene plants. These control applications have
increased process efficiency and plant capacity. The benefits of these
controllers have been well documented and the subject of numerous
papers presented at the Ethylene Producers Conference. By now, the
technology is well understood and widely accepted. The only obstacle to
further implementation and better maintenance is the cost and the
majority of the cost lies in the model identification step.

Traditionally, the Benzene Distillation process model is derived through the

open-loop perturbance of the process. This procedure is known as a step-
test and is part of any model based control project implementation. The
steptest is a manpower intensive process that often requires around-the-
clock engineering coverage and can last for several weeks. The duration is
dictated by the quality of the test procedure, which dictates the number of
perturbances required and by the duration of the process dynamics, which
dictate how frequently the tester can impart a pertubance. The cost of the
test is dictated by the duration and consists of the obvious manpower costs
and the opportunity cost of sub-optimal operation of the process, during the
test.
Furthermore, with cost cutting efforts of the process industry, in recent
years, the engineering staffs have been significantly reduced. Some
engineering staffs have been reduced to the point where in-house
steptesting is no longer possible. Therefore, new step-tests have to be
contracted to outside firms, which further increases the cost of step-testing
and thereby, raises the hurdle for new implementations and re-testing of
existing applications.

Although it is relatively easy to calculate potential benefits for a new

application, for an existing application, it is difficult to assign the
percentage of the opportunity loss due to performance degradation,
resulting from model inaccuracy. Because of the cost considerations,
existing applications are often not retested, for model improvement, until
the controller performance deteriorates to the point where the application
is turned off. Besides the existing multivariable control (MVC) applications,
there are a number of other potential applications where the economics are
less compelling but can still yield significant benefits. To eliminate the
efficiency degradation and to improve the cost effectiveness of the
economically less compelling applications we need to reduce the cost of
model identification. With smaller engineering staffs, we need to reduce the
manpower requirements for MVC implementation, which will also serve to
reduce the cost.

Application

One possible procedure for control model identification involves:

1) Building a first principles dynamic model of the process,

2) Perturbing the model rather than the process,

3) Building the control model for an MVC application from that data,

4) Commissioning the MVC controller,

5) Closed-loop testing the process for model refinement.

This procedure was implemented for a new MVC application on a
Benzene/Toluene fractionator at Huntsman, LLC ‘s Aromatics and Olefins
plant in Port Arthur, Texas.

The Benzene/Toluene fractionator is part of an aromatics extraction unit.

The aromatics are extracted from an aromatics rich stream using
liquid/liquid extraction. The benzene is separated from the aromatic extract
in a distillation column, shown in figure 1. The light and heavy key
components are benzene and toluene, respectively. The relative volatility
for these compounds is approximately 2.5, making this a separation with a
moderate reflux to feed ratio required. This means that incremental
benefits of better control are moderate and justifying a major control
project for such an application can be a challenge. But, the benefits of better
control are not insignificant, especially with the high cost of energy, in
recent years.

As shown in figure 1, the feed entering the column is controlled with a flow
controller. Pressure control is used to remove any light hydrocarbons that
are present in the extract. The benzene product is withdrawn from the
distillation column as a sidedraw product. The benzene flow is set by a
control tray temperature controller, which resets the sidedraw flow
controller setpoint. The toluene and heavier aromatics exit the column as
the bottoms stream that is controlled with a bottoms sump level controller.
Reflux is controlled with a reflux drum level controller resetting a reflux
flow controller set point. The heat input, to the column, is controlled with a
reboiler steam flow controller. With this base level control scheme, the
column operator has three Manipulated Variables (MV’s) that he can set. He
sets the setpoint for the feed flow controller to match the extract production
from the liquid/liquid aromatics extractor. The temperature controller
setpoint is set to achieve the desired benzene recovery and the reboiler
steam flow controller setpoint is set to achieve the desired benzene product
purity.

This is a fairly simple system but it is a control challenge. Due to the fact
that it is a fairly easy separation (low cost) but with disparate values of the
two products, it is a high recovery as well as a high product purity
separation. This high purity of both major products means that this is a
highly non-linear system. This implies high process sensitivity, at the
constraints. The further one goes from the constraints, the lower the
sensitivity and therefore, the easier the control. With low energy penalty
for departure from constraints and easier control, it is easy for a human
operator to drift away from constraint. This is where the automatic control
opportunity lies.

Huntsman began a project to recover that energy cost in December of 2001.

The first step was to develop a representative steady-state simulation of this
distillation column. An existing Aspen Plus[1] model was used. The model
was tuned to match a set of process data. The equipment geometry data
was entered and the simulation imported into Aspen Dynamics[1]. The base
level controllers were added to the simulation and tuned. There was no
effort made to duplicate analyzer lags and deadtimes, at this point in the
process. To demonstrate the non-linearity of the system, figure 2 shows the
steam rate required as a function of the amount of toluene in the benzene
product, at constant feed rate and control tray temperature. The steam rate
is expressed as a percentage of the steam required at the average operating
point and the toluene concentration in the benzene product is shown as a
fraction of the Huntsman constraint on that impurity.

This figure clearly demonstrates the value of running closer to constraint

and the shape of the curve shows the diminishing returns, at the higher
impurity levels.

This data was used to calculate process gains of toluene in the benzene
product to steam rate as a function of the benzene product composition in
figure 3. This figure shows the high process sensitivity at the higher
impurity levels and demonstrates that the process gain is very much
dependent on where the process is tested. Of course, the point at which the
model must have the highest degree of accuracy is at the constraints. From
figure 3, it should be clear that it would be difficult to conduct a practical
test at the constraints because of the high process sensitivity. The high non-
linearity implies that the test must be carried out in a very narrow band of
controlled variable (CV) values but with a real process, the test must impart
sufficient CV variation to extract the true process dynamics from the noisy
process data. Testing a dynamic simulation, with no noise, means that a
very small perturbance can be imparted and an accurate process gain
measured at the constraint.

To determine the right set of constraints, the Aspen Plus[1] steady-state

simulation was used to calculate the optimum operating point. The
dynamic simulation was moved to the optimum point, and was perturbed.
The data from Aspen Dynamics[1] was exported to an ASCII text file that
was manipulated to resemble an AspenTech DMCplus [2] data collection
file. That file was imported into the DMCplus[2] process model
identification software. The DMCplus [2] identification was trivial since the
data contained no noise and had perfect pure step responses. At this point,
analyzer deadtimes were estimated, by inspection of process data. This time
was added as pure deadtime to the process models. The resulting model is
presented as figure 4.

The controller configuration file was built and the controller tuned. With
this procedure, there was a high degree of confidence in the process gains,
but a very low confidence in the process dynamics. Therefore, the
controller tuning was far less aggressive than normal.

The controller commissioning was extremely smooth with minimal on-line

tuning required. In fact, the very first evening, the controller was left on-
line without engineering coverage. The majority of the benefits were
achieved on the first day of commissioning and amounted to a 15%
reduction in steam usage for the same feed rate. The control performance
was such that the plan to conduct the closed-loop testing was abandoned.
However, over time, there were periods where the controller under-
performed and was taken out of service. Therefore, the closed-loop testing
was performed. The process dynamics were identified from that data and
the new model, overlaid on the original model, is presented as figure 5. The
new model was substituted into the controller in July of 2002 and the
controller performance problems eliminated. After this iteration, any time
that the controller was found to be out of service, the explanation, from the
operators, was that no one knew why it was taken out of service and it was
immediately re-commissioned. If more than a day had passed since the
controller had been in service, the controller reduced steam consumption
by 10% to 15% as it moved the process to the appropriate constraints.

To demonstrate the benefits, the service switch position is plotted with the
reflux to feed ratio for the past 3 years, as figure 6. The only manipulation
of the raw data was to remove the process unit downtime periods. For the
purpose of this figure, the reflux to feed ratio was normalized by dividing
the actual ratio by the maximum reflux to feed ratio for this period.

February 21st, 2012 in Design Distillation System, Operations of Distillation

System