1. Ingeniería

Implementing Technology ❚❚❚❚
Feeder Accuracy And
Design: Critical Parameters
In Continuous Pharmaceutical Operations
By Sharon Nowak
Global Business Development Manager: Food and Pharmaceutical,
K-Tron International Inc.
T
The manufacturing sector of the pharmaceutical industry is currently undergoing a radical change. The need
to improve process and product quality while controlling overall process costs is growing. Continuous processing offers a chance to achieve these goals. The
implementation of PAT (process analytical technology)
tools further aids in the support of the continuous operation, as the tools give real-time indication and assurance of product quality throughout the process.
A variety of batch processes in the industry are currently undergoing continuous evaluation, including
continuous direct compression or blending, continuous
wet and dry granulation, and continuous granulation
via hot-melt extrusion. At the heart of each of these continuous operations is the feeder which is used to proportion raw ingredients into the process. Feeder technology and accuracy is paramount in the overall continuous process.
Feeder Accuracy Defined
To fully define feeder accuracy, it is necessary to
address three separate and distinct areas of feeder performance: repeatability, linearity, and stability.
Repeatability reports how consistent the feeder’s discharge rate is at a given operating point; linearity assesses how accurately the feeder discharges at the requested average rate over its full operating range; and stability gauges performance drift over time.
Repeatability is the performance statistic most familiar
to feeder users. It quantifies the short-term level of consistency of discharge rate. Repeatability is important to
quality assurance because it measures the expected
variability of the discharge stream and of the product
itself. The repeatability measurement is made by taking
a series of carefully timed, consecutive catch samples
from the discharge stream, weighing them, and then
calculating the +/- standard deviation of sample weights
20
The Pharmaceutical Solutions Update
expressed as a percentage of the mean value of the
samples taken.
While repeatability measures variability of flow rate,
linearity reports how well the feeder delivers the
desired average rate throughout the feeder’s operating
range. Perfect linearity is represented by a straight-line
correspondence between the set point and the actual
average feed rate throughout the feeder’s specified turndown range from its designed full-scale operating point.
To perform a linearity measurement, several groups of
timed catch samples must be taken from the feeder’s
discharge stream. Typically, 10 consecutive catch samples are obtained and weighed at each of the following
flow rates: 5%, 25%, 50%, 75%, and 100% of full scale.
(The smallest tested flow rate should be at the feeder’s
maximum turndown.)
A perfectly performing feeder is worth little if it cannot maintain its performance over the long haul. Many
factors can potentially contribute to performance drift,
including feeder type, control and weigh system stability, the handling
characteristics
and variability of
the material, the
feeder’s mechanical
systems
maintenance,
and the operating environment
itself. Drift is
detected by calibration checks
and is typically
remedied by a Feeder technology and accuracy is paramount in overall continuous evaluation,
simple weight including continuous direct compression or
span adjustment. blending, and continuous wet and dry granuThe user will ulti- lation process.
www.pharmaceuticalonline.com
Implementing Technology ❚❚❚❚
mately determine the appropriate frequency of
calibration checks based on operational experience. For accuracy requirements in the 1% to
5% range, volumetric feeders will usually suffice, while gravimetric feeders are used for performances in the .25% to 1% range.
way of knowing
the out-of-control
condition. Since
the feed rate in a
volumetric feeder
is purely a function of speed, the
feeder and the
Gravimetric Versus Volumetric
process
below
Feeding For Continuous Operations
By definition, gravimetric feeders measure the
have no way of
flow’s weight in one fashion or another and
detecting
this
then adjust feeder output to achieve and mainupset condition.
tain the desired set point. Volumetric feeders
Oftentimes, even
do not weigh the flow; they operate by deliv- Loss-in-weight feeders are typically either the use of level
ering a certain volume of material per unit mounted on weigh scales or suspended from load sensors in the feed
time from which a weight-based flow rate is cells and offer good material containment.
hopper may not
inferred by the process of calibration.
alert the process of
The feeders described below incorporate the twinthis upset in a timely fashion. Most gravimetric feeders
screw design. Twin screws are often used in the pharcan automatically detect and alarm to these conditions.
maceutical industry because they are much better suitUnderstandably, the highest accuracy in a screw feeded for non-free-flowing or cohesive materials. The use
er can be attained on free-flowing materials that fill the
of two co–rotating screws allows for a constant “self
screw consistently and whose density is reasonably conwiping” effect of one screw against another, thus helpstant regardless of hopper level. Volumetric screw feeding to maintain constant flow without buildup in the
ers represent an economical solution to many process
screw tube.
feeding applications, as opposed to the higher costs of
equivalent gravimetric or loss-in-weight designs.
Volumetric Feeding
As mentioned earlier, volumetric feeders operate by
delivering a certain volume of material per unit time.
The Loss-In-Weight Principle
The most popular type of gravimetric feeder used in
Flow rate changes are accomplished by altering screw
continuous pharmaceutical processes is the loss-inspeed. In the case of a screw feeder, three factors affect
weight feeder. Loss-in-weight feeders directly measure
volumetric screw feeder accuracy: the consistency of
and control the flow rate to the process. Loss-in-weight
delivered volume per screw revolution, the accuracy of
feeders are typically either mounted on weigh scales or
screw speed control, and material density variability.
suspended from load cells and offer good material conTypically, volumetric feeders are open-loop devices
tainment.
and cannot detect or adjust to variations in the materiA loss-in-weight feeder consists of a hopper and a
al’s density. Due to the
feeder that are isolated from the process so the entire
open-loop concept, headsystem can be continuously weighed. As the feeder disload variations and matericharges material, system weight declines. The speed of
al buildup on the feed
the metering device is controlled to result in a per-unitdevice change the volumetime loss of system weight equal to the desired feed rate.
per-revolution relationship,
A typical loss-in-weight feeder controller adjusts feeder
throwing off calibration
speed to produce a rate of weight loss equal to the
without any outward sign.
desired feed rate set point.
Gravimetric feeders autoIn the plot of weight vs. time, feed rate set point is repmatically detect and adjust
resented as a downward sloping line. The negative
to these conditions. In
cases of screw feeding of
mathematical slope ( W/ T) indicates the desired loss
cohesive materials, it is
of system weight per unit time. The feeding cycle begins
possible in volumetric
with a fully loaded hopper where weight is at a maximode to have relatively no
mum. As feeding proceeds, the measured weight is conmaterial discharging while
tinually compared against the set point line’s target
Volumetric feeders operate by
the screws are running.
weight. Any difference between the two values triggers
delivering a certain volume of
Similarly,
flood-through
can
a change in feeder speed. For example, if an overfeed
material per unit time. Flow rate
condition occurs due to an abrupt increase in material
changes are accomplished by also remain undetected
since the feeder has no
density, sensed weight falls below desired (set point)
altering screw speed.
22
The Pharmaceutical Solutions Update
www.pharmaceuticalonline.com
Implementing Technology ❚❚❚❚
weight, triggering
a reduction in
screw speed to
return to the se
point
value.
Additionally, since
the
integrated
error associated
with the overfeed
is known, screw
speed may be further reduced to
immediately and
A typical loss-in-weight feeder controller adjusts
precisely compenfeeder speed to produce a rate of weight loss
sate
for the overequal to the desired feed rate set point.
feed
condition.
The opposite occurs with an underfeed condition.
Refill Operations For
Continuous Processing
In continuous processes, the design of the refill device
is also critical to the feeder’s performance. The design
of this upstream component, as well as the additional
options available for overall feeder design, will affect
not only process efficiency but also operation, safety of
the process, and end product quality.
When the feeder hopper’s charge of material becomes
low, it must be resupplied. However, since feeding
accuracy relies on the ability to continuously weigh the
feeding system, direct gravimetric control during this
period is not possible. Even though refill may occur
quickly, it is important that feeder performance be
maintained during refill.
Lacking any basis for gravimetric control during these
brief but periodic refill phases, flow control is achieved
volumetrically. Traditionally, a constant metering — a
speed corresponding to the metering speed associated
with gravimetric control just prior to entering the refill
phase speed — is maintained throughout the refill
phase. If, for example, metering speed averaged 60 rpm
just prior to the system sensing the need to refill the
supply hopper, screw speed would be maintained at
that 60 rpm for the duration of the refill operation. After
refill is completed, material has settled, and the feeder
senses an appropriately declining system weight, the
feeder is returned to gravimetric operation where metering speed once again becomes the parameter of control.
To minimize feed rate errors during refill, another
approach is required that enables metering speed to be
gradually lowered during refill to precisely counterbalance the effects of increasing material density occurring
within the metering zone as hopper weight increases.
The slower rate is determined by storing an array of
indices called feed factors in the controller’s memory.
These values correspond largely to material density and
24
The Pharmaceutical Solutions Update
its mechanical behavior within the feeder, and are computed during the entirety of the gravimetric feeding
cycle. As net hopper weight declines, the controller also
determines and stores a set of up to 100 feed factors,
each of which is an index of the average density of
material discharged at the hopper weight associated
with the feed factor. For example, a low feed factor indicates that a higher number of screw revolutions were
required to discharge a given weight, implying a
reduced material density. Then, on the basis of sensed
hopper weight at each array point during refill, material density within the metering zone may be inferred,
and a metering speed corresponding to its feed factor
array value may be invoked. In this way, gravimetric
feeding accuracy during the brief refill may be maintained.
Typical components of a refill system for a feeder
often include a vacuum loader located above the refill
hopper. NOTE: An integrated refill system which controls the operation of the loader and the refill request of
the feeder is a critical requirement to control overall
process accuracy.
Other Design Parameters To Consider
In addition to choosing the proper feed technology,
care must also be taken in the overall configuration and
design of the feed system components. For example, in
the case of loss-in-weight feeding, in-plant shock and
vibration act to corrupt the weight measurement,
destroying the basis for feed rate control. Flexible connections and the possible use of shock mounts help isolate the feeding system. In addition, both the weighing
and control systems must be designed to discriminate
between meaningful weight readings and the spurious
forces associated with residual shock and vibration.
In the case of feeder design for the pharmaceutical
industry, overall sanitary design and cGMP construction,
as well as safety considerations for active ingredient
handling must be considered. Additionally, the material
being fed as well as the cleaning materials that will be
utilized should be considered. When working with
potent compounds, consideration must be given to the
containment of the feeder, as well as the method of
delivery of the material both to the feeder and from the
feeder to the process below.
In many installations, space and ergonomics of design
are important factors in the overall feeder design chosen. Such options as slide bases or diverter valves at the
feeder outlet for online feeder calibration can be included. In addition, alternate portability designs, such as
modified feeder mounts and the use of lift mechanisms,
should be considered, in order to provide for easier
access during all phases of production, maintenance,
and cleaning. When selecting any feeder technology,
these requirements should be addressed with the feeder manufacturer in question. ●
www.pharmaceuticalonline.com