|
The Design of Loss-in-Weight
Feeders
Guest article by Mike Page of Rospen Industries
 Forward
This is the second of two papers from
Rospen Industries, a UK based, global supplier of weighing and
metering screw feeders and relates to the design of continuous
loss–in-weight and batching loss-in-weight type feeders. Together,
with the necessary considerations for the sizing of the associated
loss-in-weight feeder hopper, this paper also highlights some of the
typical problems and possible solutions confronting a wide spectrum
of industries using this type of equipment.
Terminology
Whatever means are employed to
actually control the powder, certain basic fundamentals need to be
assessed and understood before a solution can be arrived at.
Firstly, the terminology used to describe aspects of the equipment
is universal and whilst this was given on the first paper it is
worth summarizing again and is as follows:
Diameter of Screw: This is the
nominal outside diameter of the screw not the diameter of the tube
in which it runs, nominal, because a 40 mm screw is in fact 38 mm
diameter.; a 20 mm screw is 19 mm diameter etc. These differences
only occur because the nearest available stainless steel tube
schedule requires selection of these dimensions to give a proper
running clearance.
Pitch of Screw: The dimension from
leading face to the next leading face of the screw. In metering
screw feeder terms the pitch dimension relates to the nominal screw
diameter, applicable on screws up to 100 mm diameter. Above this
size, the pitch will be smaller than the diameter but is still
referred to as full pitch.
Box Loading: This refers to the level
of material in the screw tube/casing and is usually quoted in %
terms i.e. 50% etc. When box loading is related to the screw pitch
then fractional terminology i.e.1/2, 1/3 is invariably used. For
reference, Rospen apply 100% loading. However, for long feed
applications the screw may require an expanding pitch to reduce the
box loading in the tube.
Volumetric Feeding: Strictly based on
a screw of a known diameter and pitch, metering powder at a
specified speed. It must be noted that that any variation of
material bulk density will have a direct effect on the ultimate feed
rate. (A separate design paper is available, which covers the design
principles)
Gravimetric or Loss-in-Weight:
A feeder usually identical in design to the volumetric feeder but
mounted to a weigh platform to measure the weight lost from the
feeder at intervals of time.
Principles Of Loss-in-Weight
Feeding (gravimetric feeding)
With regard to the design of a weigh
screw feeder there is a cardinal rule that if a powder or product
cannot be handled with reasonable accuracy volumetrically,
installing a weighed system will not solve the problem and could
make matters worse.
When reading the following notes it
will become apparent that the feeder hopper capacity is an integral
aspect of the system, together with the correct filling of the screw
and its design, (which has been covered in the earlier paper). These
points will provide a correctly sized unit to ensure the required
throughput can be met and accurately weighed, with minimum time in
volumetric mode during the refill period.
Set out below is a section relating
to the advantages of a weighing system followed by an example of
calculations used for hopper sizing and finally a set of key design
points for both, continuous and batching loss-in-weight systems.
Advantages of a Weighing System
- Feed rates can be maintained
within tighter tolerances even with variations in bulk density,
which can occur with different suppliers of the same powder.
- The controls can be supplied by
any commercially available PLC manufacturer with an HMI
configured to show the process, whether it be a single or
multiple machine operation.
- The ability to set high and low
level alarms on the feed rates, which provides a fail safe to
the process and prevention of out of specification material
mixes.
- Use of a totalizer system in the
software, which will log total quantity of material fed on any
time basis (shift, day or month) to assist with the inventory
control of the process.
- An option to remotely
interrogate the control system via an Ethernet or Profibus Comms
module (depending on PLC supplier).
- Remote control of the system to
set rates using the same route.
Sizing of Continuous
Loss-in-Weight Feeder Hopper
Take maximum feed rate kg/hour and
divide by the loose fill material bulk density in kg/m3 This will
give cubic meters/hour throughput.
- Divide by 6 (ideally a 10 minute
run in gravimetric mode).
- Multiply above by 1.3 to allow
for the ullage above the net material level.
- Select nearest size of standard
hopper above this.
Example
Size the hopper for 500 kg/hour
continuous duty with material at 800 kg/m3
1) 500/800 = 0.625m3/hour
2) 0.625/6 = 0.104m3
net refill required every 10 minutes
3) 0.104 x 1.3 = 0.135m3
Select 150 liter standard hopper.
Continuous Loss-in-Weight Feeding
Hopper refill rates are critical and
ideally will be 10 times the output rate of the feeder. When a
feeder is re-filled it moves to volumetric mode based on the last
received signal when in loss-in-weight. Until the weigh platform
senses the high level of material in the hopper it will stay in
volumetric mode.
After the refill it will, within 3
seconds, re-learn the loss-in-weight signal and react accordingly.
Obviously the smaller the refill rate
with the screw still discharging product, the longer the volumetric
stage and the benefits of loss-in-weight are reduced.
Any size of feeder up to 150litre
capacity hopper can be mounted to a weigh platform. Larger hoppers
will require to be suspended from a weigh frame to provide stability
of the system.
In both platform and weigh frame
configuration, it is usual to fit three load cells on the milking
stool principle to ensure even loading of cells.
As with all weighing systems the main
problems are caused by site vibration particularly if it is of a low
frequency high amplitude characteristic such as locally sited ball
mills or en masse vibrating tray conveyors with heavy reaction
bases.
The fitting of anti-vibration mounts
can help under the feeder/weigh frame in limiting the effect. The
control systems can also be set to filter out some of the excessive
signal noise caused to the load cells.
On particular large rate
loss-in-weight feeders, where there is a requirement to control say
20 Te/hour of product with a refill rate of 200 Te/hour, it becomes
on most installations, impossible. Even direct discharge from a silo
may not keep up with this rate.
The alternative is to run a
continuous flow meter device whether screw or belt with a volumetric
pre-feeder usually a screw.
The two are run in closed loop
control to maintain the set rate with the added advantage that the
pre-feeder and hopper, because it is volumetric, does not require
the same high rate of refill as the gravimetric feeder. Usually only
two to three times the output rate.
The feeder hopper is fitted with high
and low level probes to control refills in the conventional sense.
While this solution is invariably
more expensive due to the number of machines employed, it provides a
stable and sure method for the control of large feed rates. The
pre-feeder can also be a vibrating tray if the product to be handled
is sensitive to compaction e.g. cereal flakes or IQF fruit.
Batching Loss-in-Weight Feeding
Uses the same controller as
continuous to measure the loss of material from the feeder but is
set to switch off when a pre-determined target weight is reached.
The operating mode is usually two speed i.e. fast /trickle with the
occasional requirement to jog the drive for part revolution of the
screw in high accuracy requirement.
While it is ideal to batch in one
operation, if it is impossible to get a large enough hopper in
place, multiple runs can be made to make up the batch but refill
rates may not be so critical as a continuous loss-in-weight system
due to the batch cycle time of the process.
Problems And Possible Solutions
Problem: Feed rates keep
dropping causing screw speeds to increase to alarming levels and
then suddenly revert back to where it was. Usually on a cyclical
basis.
Solution: Material is building
up within the pitch of a solid screw and reducing the carrying
volume. It reaches a point where it cannot sustain its own weight
and breaks away leaving the pitch clear to fill properly again.
Replace with a wire screw to cut down surface area for build up.
Problem: Cohesive/Sluggish
Powders.
Solution: Do not attempt to run
at maximum speeds and size the screw and pitch to reflect this. Use
either twin screws or a wire design or combine both.
Problem: Friable Granules
Solution: Obviously slow speed
with increased tube diameter with a wire or solid screw will be
acceptable. Alternatively, use a vibratory tray mounted to a weigh
platform.
Problem: Segregation of mixed
powder with widely varying bulk densities.
Solution: Try not to use
vibration on the hopper and minimize the agitator effect if possible
to a 2 blade intermittent action.
Problem: Extreme pulsing effect
from the tube outlet causing loss in weight alarms.
Solution: Usually this becomes
more apparent at reduced screw speeds, therefore, select a screw
size to provide a sensible speed range or fit a cross wire sleeve to
break up the materials extruded structure. In extreme cases a small
air vibrator can be mounted to the tube to provide the same effect.
Problem: Flushing – usually
caused by excessive air entrained within the product by a previous
operation.
Solution A: If sufficient time
delay cannot be provided within the feed hopper to condition the
material with vibration then the use of a large fabricated rotary
vane, similar to a rotary valve, can be considered as a replacement
for the agitator and located in the exact same spot over the screw.
This has been successfully retro-fitted on several occasions.
Solution B: Fit close tolerance
discharge tubes and/or increase the number of pitches of the screw
by fitting a twin start or double start blading. The ultimate fall
back is the twin screw feeder but physically constraints usually
prevent this from being retro-fitted. It is important to bear in
mind the drive for the screw on a twin is quite different from the
single and requires a splitter gearbox to transfer a single input
into a double output. The baseplate is, therefore, considerably
longer.
Problem: Cannot gain sufficient
height to feed into the client’s process from the infeed position.
Solution: Incline the screw.
Invariably this results in an increase in feed centers so the motor
power rating must be increased to compensate. With volumetric
feeders just provide a frame to support the feeder, but in the case
of gravimetric feeding ensure the feeder is suspended from a
horizontal weigh frame to improve the stability and accuracy of
weighing.
Another point to consider on inclined
screw applications is that the face angle of screw as it inclines
becomes less effective in feeding. To overcome this, size the screw
pitch to obtain the best angle possible otherwise material will fall
back over the blade giving greater slippage and less efficient
conveying.
Conclusion
The design criteria and advantages of
the use of Loss-in-weight feeder system is laid out in the preceding
article but with this must be the answer to the most asked question,
how accurate is it? The answer, 0.5 to 1.0%.
For more information contact:
Mr. Mike Page
Senior Sales Engineer
Rospen Industries
Oldends Lane Industrial Estate
Stonehouse, Gloucestershire
United Kingdom, GL10 3RQ
Email:
mikepage@rospen.com
Web site:
http://www.rospen.com/
Help others by posting your comments,
suggestions and experiences with bulk solids feeding or any other
materials handling concerns you may have on our
On-Line
Help Forum.
For past Ask Joe ! Articles, visit
the
Ask Joe! Archived Articles.
Guest articles for the Ask Joe!
Column are always welcome, for more information please contact Joe
Marinelli directly at his email address:
joe@solidshandlingtech.com.
|
