International Journal of Scientific Advances
ISSN: 2708-7972
Volume: 2 | Issue: 5 | Sep - Oct 2021 Available Online: www.ijscia.com
DOI: 10.51542/ijscia.v2i5.25
Arduino based control of the Food and Water Conveyance
Systems of a Refractance Window Dryer
Raymonds Mutumba1, Julia Kigozi1*, Peter Tumutegyereize1,
Shaffic Ssenyimba1 and John Muyonga2
1Department
of Agricultural and Biosystems Engineering,
Makerere University, Kampala, Uganda
2Department
of Food Technology and Nutrition,
Makerere University, Kampala, Uganda
*Corresponding author details: Julia Kigozi; jbulyakigozi@yahoo.com
ABSTRACT
A refractance window dryer with a 14.5kg/hr throughput capacity was developed to effectively dry food
product of 3mm on the conveyor belt. For efficient dryer performance an automated system for the
conveyor belt movement and water conveyance system was designed. The automated system comprised
of an ARDUINO centered control system, an arrangement of sensors, water pump and the conveyor
motor. A computer program was written in Arduino environment, successfully compiled and uploaded
on to the controller board to process all commands. The system was fi rst simulated successfully in ISIS
Proteus environment and connected onto a bread board for testing before attaching the motor onto the
main circuit board. Performance tests done at 85°C revealed that there was no movement of the belt as
temperature built steadily from 31.19°C until it reached a temperature of 92.0°C in the boiler. The
maximum recorded water temperature was 98.06°C and the system had an operating range of 95±3°C.
Achieving this led to an automated food conveyance system that was reliable an d ensured high product
quality. The Arduino based system worked well and is recommended for the refractance window dryer
and can be up scaled to a bigger similar machine.
Keywords: dryer; automation; Arduino; food conveyance; water conveyance
INTRODUCTION
Food drying is one of the most important processes in the
food industry, and it demands various levels of energy, to
produce commercially dried food products of high
quality (Mujumdar, 2007). To address the challenge of
heat sensitive fruits, non-thermal technologies have been
used which include ultrasound, pulsed electric fields and
ultraviolet radiation which do not generate heat within
the product (Raso and Barbosa, 2003) hence avoiding
case hardening. Such technologies though, have a high
capital cost of design and operation attached to them
(Onwude et al., 2017) as compared to the refractance
window dryer (RWD). In the design of an RWD, the food
products are dried over a bed of hot water and conveyed
over the hot water using a mylar conveyor belt. The food
conveyance system comprises of all mechanisms and
components of the dryer responsible for moving food
product from the dryer inlet to the outlet. It is mainly
comprised of the conveyor belt onto which food is placed,
a conveyor motor that drives the belt and an idler
conveyor pulley around which the belt goes. Automation
the technology by which a process or procedure is
accomplished without significant human assistance
(Groover, 2008), is applied to equipment for increased
efficient performance. To automate a process, power is
required both to drive the process itself and to operate
the program and control the system. Automation of the
food conveyance system of the RWD dryer would require
the development of a central controller system through
which all signals and commands that operate the system
are passed (Nurus, 2017). The controller receives signals
from the food detection mechanism comprising an
arrangement of infrared sensors that sense the presence
or absence of food on the belt and sends a response to the
controller board. The length and speed of the conveyor
mechanism determines both the residence time and
moisture content of the final product (Shirinbakhsh,
2017). The reliability, safety factor and increased
efficiency of automated food conveyance systems make
them more preferred in both small- and large-scale
industrial applications. An effective food conveyance
system would need to have a quick response to
commands and every machine designed is required to
have its own system depending of the unique operating
parameters.
MATERIALS AND METHODS
The RWD synchronized conveyor belt and water
pump systems.
The RWD food conveyance system comprised of a
temperature sensor which reads water temperature in the
reservoir and sends signals to the controller that triggers
the variable frequency drive as illustrated in FIGURE 1. The
VFD is responsible for controlling the conveyor motor
speed to any speed required during operation.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
810
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 1: The food conveyance system components
The DS18B20 immersible temperature sensor was selected
because of its high degree of accuracy, tolerance to high
temperatures, and quick response to temperature
variations. The 16X2 LCD and the Arduino Uno controller
board were selected because they easier to program and are
readily available in the local market. Compared to other
programming boards like Raspberry Pi, Orange prime and
OdroidXU4, Arduino boards are easier to interface with
sensors and can provide on board storage which makes
them preferable for this project. The food conveyor belt was
programmed to move when the temperature in the boiler
reaches the desired ranges.
Design of the DS18B20 temperature sensor circuit
The sensor circuit was designed as shown in FIGURE 2. The
temperature sensor measuring water temperature is
connected to the controller board. The sensor is designed
to be powered with 5V connection pin from a standard
Arduino controller board with the VDD terminal connected
to an external supply of +5V. A liquid crystal display (LCD)
was embedded in the circuit to allow for display of the
measured parameters.
FIGURE 2: DS18B20 temperature sensor circuit connections
The following circuit diagram shows how the LCD was
connected with the Arduino module. From the circuit
diagram in FIGURE 3, the RS pin of the LCD was connected
to pin 12 of the Arduino and the R/W pin of the LCD was
connected to the ground. Pin 11 of the Arduino was
connected to enable signal pin of LCD module. The LCD
module and Arduino module were interfaced with the 4bit mode hence there were four input lines; DB4 DB5 DB6
and DB7 of the LCD.
This circuit required fewer connection cables and utilized
the most potential of the LCD module (Circuits Today,
2020). A 10K potentiometer was used to adjust the
contrast of the display, 560Ω resistor R1 limits the current
through the back-light LED and the LCD was powered
through the +5V pin provided on the board. A computer
program for the LCD to display LM35 temperature was
written using Arduino software and uploaded on the
board.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
811
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 3: 16x2 Liquid crystal display pin arrangement with Arduino
The Variable Frequency Drive_ VFD circuit design
Selection criteria for the VFD included compatibility with
230/240V single phase power source and ability to drive
up to 3HP AC motors. The drive was designed with a range
of 0-50Hz for min-max speed operation. The VFD main
external circuit was wired as illustrated in FIGURE 3 where
R, S, T were VFD input terminals from the main power
supply and U, V, W were VFD output terminals to the
motor. The VFD was powered from an AC 230V/ 240V to
through pins RLI and SL2 and earthed at E. A Brake Resistor
and Brake Unit was used on the VFD to reduce the
deceleration time of the motor. A magnetic contactor (MC)
was embedded in the power input wiring to cut off power
quickly and reduce malfunction when activating the
protection function of AC variable frequency drives. The
screws of the main circuit terminals were fastened to
prevent sparks that can be caused by the loose screws due
to vibration. To permanently reverse the direction of motor
rotation, any of the two motor leads could be switched over.
FIGURE 4: Variable frequency drive main circuit connection
The frame and mounting of the VFD were designed in such
a way that it allowed for enough airflow through the VFD
to avoid over heating of the device by leaving a clearance
large enough for air circulation. The total volume
requirement of the mounting was calculated from
Equation 1
Volume; 𝑉 = (𝐶 + 𝑊 + 𝐶) × (𝐾 + 𝐻 + 𝐾) × (𝐿)
Equation 1
= (2𝐶 + 𝑊) × (2𝐾 + 𝐻) × 𝐿
Where W is the width of the variable frequency drive, H is
the height of the variable frequency drive and L is the
breadth of the mounting. K and C are clearances
horizontally and vertically between the VFD and mounting
all-round respectively. The VFD frame was made out of
aluminum since it is able to withstand the high
temperatures. A wooden backboard was made onto which
the VFD mounting will be secured by use of screws that run
through the VFD holes into the back board. The VFD was
mounted vertically on a flat vertical surface (mounting) by
screws, other directions were not allowed.
The VFD generated heat during operation hence a
clearance of 50 mm minimum was allowed around the unit
for heat dissipation since heat sink temperature may rise
to 90°C when running (Delta Electronics, 2008)
Design of the water circulation system
The RWD water conveyance system comprised of a relay
module which was triggered by the controller board to
switch the motor on or off. The relay module is powered by
an adaptor which send an electric field that magnetizes the
cores of the relay. The water pump was connected to the
output cable and the relay as shown in
FIGURE 5. Automation of this system was done to ensure
that the water pump pumps boiling water only when it has
reached the desired 95°C. The red power wire was
identified in the cord leading to the water pump and a cut
was made. The wire side leading to the water pump was
connected to the NC terminal of the relay and the side
leading to the plug connected to the COM(C) terminal
hence the relay was on the hot side and the current from
the plug was switched before it reaches the water pump.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
812
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 5: The water conveyance system circuit diagram
The relay module could be powered by an independent
+5V adaptor but in this case, it was powered by the
Arduino board pin2. The Vcc terminal on the relay module
was connected to the +5V on the Arduino board and the
ground terminal of the relay connected to the GND on the
Arduino. The signal terminal of the relay module was
connected directly to pin7 on the Arduino board to receive
signals and to trigger the relay to flip.
Computer program design for the food and water
conveyance system
The flow diagram in Figure 6 shows a logical flow of
information and a guidance as to how the computer
program was written.
The program was written using Arduino software and
uploaded on to the Arduino board via USB drive. A
program script was written to receive the analog signal
from the microcontroller and respond by switching on the
VFD when temperatures reach the desired value. A
separate program script was written for the LCD to display
the different temperature values as read by the sensor. A
program script was written to receive the analog signal
(analogRead) from the DS18B20 temperature sensor and
convert it into digital signal to be read by the
microcontroller. The microcontroller has been programmed
to command the relay to close and switch on the conveyor
belt only if the water temperature has raised to more than
95°C (Temp>95°C)
FIGURE 6: A data flow diagram for the water pump and motor control computer program
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
813
International Journal of Scientific Advances
A separate code was written and uploaded to the Arduino
board for the LCD to display the different temperature
values as read by the DS18B20 sensor.
Performance testing of the synchronized water and
food circulation system.
The objective of the testing was to evaluate the dryer
performance in terms of drying time, power consumption,
temperature along the belt and product moisture content.
The system comprising a conveyor motor, water pump,
temperature sensor and the IR sensors all connected to the
controller board was set up. The Controller board was
connected to the computer through a USB cable to enable
communication and information flowing from the control
board to be read by the computer. The set up comprised of
PLX-DAQ computer software was used to convert the
ARDUINO software performance results into Excel results
that could be tabulated and analyzed. The system was
switched on the temperatures in the boiler recorded as
well as the desired temperature set by the knob, water
pump discharge and conveyor belt speed. According to the
system calibration, the dryer is able to dry different food
products but mango slices were used for this specific
performance test. The drying residence time was
calculated as the difference between the beginning of
drying and the time at the end of drying.
ISSN: 2708-7972
The optimum drying temperature for the selected product
is 95°C but the performance testing was done at 85°C in the
tray due to the several temperature loses in the system.
When the temperature desired temperature was reached,
the timer was started and the product surface temperature
for three different random slices was recorded using an
infrared temperature gun every after 10 minutes. The
ambient temperature and relative humidity inside the
dryer were also read using a humidity reader as shown in
FIGURE 7 and tabulated. Mango slicing was successfully
done and the 3mm thickness was maintained by
measuring the thickness of every slice using a ruler and
loaded to ensure that slices are evenly spread across the
breadth of the belt for effective drying. For effective water
temperature monitoring, the DS18B20 temperature
sensor was moved from the boiler to the drying bed to
detect temperature changes. This was done because it of
the large temperature difference between the temperature
of water in the boiler and that in the drying bed.
Temperature changes along the conveyor belt during
drying were also analysed to understand the heat
distribution during drying. Using an LCD digital
thermometer with probe, the water temperatures along
the conveyor belt were recorded at intervals of 1m from
the product application side of the tray. This was done at
different intervals of time and recorded.
FIGURE 7: Recording of product surface temperature and Humidity during drying
• Determining product moisture content
The moisture content of the mango samples was
determined by the oven drying method described by
Chaudhary (2016). Three samples were selected every
after 10 minutes, whose moisture content would be
analysed. The weight of empty dishes was recorded before
loading the fresh sample.
Samples were the weighed before and after drying in the
oven using a digital weighing scale as shown in FIGURE 8.
The samples were dried for 16 hours at 50⁰С and results
tabulated. The oven drying method for moisture content
calculation proved to be effective since all the 27 samples
selected could be tested at once hence reducing space for
errors.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
814
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 8: (A) Weighing of samples. (B) Samples placed in the oven for moisture content calculation
The moisture content was calculated accordingly from
Equation 2
𝑀𝑠 =
(W1−W2)
(W1−W3)
𝑋 100%
Equation 2
Where 𝑀𝑔 = Moisture content of the grains, W1 = Sum of
weight of wet sample and that of the dish, W2 = Sum of
weight of the dry sample and that of dish and 𝑊3= Weight
of the tray
System simulation using Proteus (ISIS)
The system was simulated using ISIS software to verify
system design and simulate them in real-time environment
which reduced the design time and cost. Proteus is a
simulation and design software used to create schematics
and electronic prints for manufacturing printed circuit
boards and simulate the circuits in real time.
Selection of the simulation software was based on the ease
with which the software is compatible with Arduino since
it requires a simple drag and drop of the written code to
run the simulation environment. After assembly, the
system was then RUN to view functionality and possible
errors. All the circuit designs described in the preceding
chapters were first simulated in the ISIS environment
before connecting them on the bread board and finally to
the PCB boards for testing.
RESULTS AND DISCUSSIONS
Results for system simulation
The computer program written and uploaded ISIS
simulation software was able to run successfully as shown
in FIGURE 9. The simulation environment showed that
whenever temperatures were increased beyond the
desired temperature, the water pump and conveyor belt
were switched on by the system.
FIGURE 9: System simulation using Proteus (ISIS) software
In the simulation, there was a delay of 3 seconds between
the time when the pump and the conveyor motor were
turned on by the system. This delay is because of the delay
line include in the void loop section of the program that is
uploaded on to the controller board. This was to ensure
that the water pump replaced the cool water in the tray
with hot water to avoid semi dried product at the delivery
end of the belt. In the simulation results, an LM35
temperature sensor was use instead of the DS18B20 to
adjust temperatures. This is because the LM35 was more
compatible with the controller board in terms of sending
commands, the results could have been similar if with the
use of the DS18B20 simulation sensor as well.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
815
International Journal of Scientific Advances
The conveyor belt circuit results
Results from the system simulation were used to verify the
computer program that was written and the system circuit
design before assembly. Basing on the simulation circuit
design, the controller board and LCD were interfaced and
the circuit illustrated in Figure 10 was connected.
ISSN: 2708-7972
The computer power through the USB cable was found
sufficient to power the control board during testing but an
adapter was soldered directly onto the board for powering
onto the dryer.
FIGURE 10: Connection of the LCD on the circuit board
FIGURE 11 shows the drier into which the controller box
marked “A” was installed. Signal lights of different colours
were installed onto the box that indicated whether a
specific device was ON/OFF. Output devices monitored
include blower, cooling and system water pump, washing
unit, conveyor belt, boiler, controller board and overall
dryer ON. To enhance further air circulation, a fan was
mounted onto the control box to avoid accumulation of
heat inside the box which would affect the motor drives
and other components installed therein.
FIGURE 11: control board installed onto the dryer for testing
The food conveyance system results were the graph in
FIGURE 12 was extracted. The belt state was programmed
to indicate and match the minimum belt speed when the belt
was off and the maximum belt speed when the belt was
turned on. Results showed that the there was no movement
of the conveyor belt (LOW) as temperature built steadily
from point A at 31.19°C until it reached a temperature of
92.0°C in the boiler indicated by point B. At 92.1°C, the
conveyor belt was turned on moving at a linear speed of 5
m/hr. At point C, the boiler temperature drastically dropped
from 95.75°C to 63.81°C which prompted the conveyor belt
to momentarily stop until the temperature built back to
92.1°C. However, this is an anomaly caused by a malfunction
in the sensor communication with the controller board.
When the temperatures momentarily drop below the
desired temperature, the conveyor belt is supposed to
continue moving. This is seen at point E when another
anomaly of sudden drop in temperature from 95.75°C to
63.81°C was recorded by the system but in this case, the
conveyor belt continued to move regardless of the low
temperatures recorded. The water temperature reached a
maximum recorded temperature of 98.06°C indicated s
point D on the graph and at this point, the temperature
started to gradually drop. The drop-in water temperature
after point D is attributed to the programing of the system
which is designed to maintain the temperature within the
desired range by switching off the water heater hence the
drop in the temperature.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
816
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 12: Relationship between boiler temperature variation and conveyor belt state
The conveyor belt did not move before the temperatures
reached the desired set temperature of 95°C to avoid loss
of product as semidried product. If the belt had moved
before desired temperature was reached, there would
have been product at the delivery end at a lower moisture
content than what is required hence product loss. The
conveyor belt was turned on by the system at a
temperature of 92.1°C, this is because it had reached the
desired set temperature of 95°C with an allowable range of
±3°C. This means that temperatures anywhere between
95°C and 98°C would trigger the belt to go on.
The variable frequency drive was connected in the
mounting and tested as illustrated shown in FIGURE 13. A
2.5hp Altivar 31 variable frequency drive was selected to
run the 2hp conveyor belt motor and an Allen Bradley
1.5hp labeled “B” was selected to control speeds for the
washing unit motor. Both drives were able to convert
power from single phase to 3phase drive since these were
the motors selected for these dryer units. The mounting
volume was calculated from Equation 1 was found to be
0.018m3. The clearance C of the mounting was 50mm on
either side of the drive to allow for sufficient air circulation
around the drive for effective cooling. The motor drive was
able to detect the Out-of-Phase-Fault and Short-CircuitFaults, and displayed them during testing.
FIGURE 13: (B) VFD testing (C) Altivar31 motor drive displaying the short circuit error during testing
The water conveyance system
For the design, two relay modules were used; HF3FA/012ZTF Hongfa relay module and a AFEBRD-SS-112LM 12VDC
15A relay. The food conveyance system was operated,
results generated and the graph in FIGURE 14. extracted.
The water pump state indicated the minimum value
(0m3/hr) when the water pump was off and the maximum
value (3m3/hr) when the water pump was turned on.
Results showed that the water pump was off during the
time when the boiler temperature built steadily from point
A at 31.19°C until it reached the set desired temperature of
92.0°C in the boiler marked as point B on the graph. At
92.1°C just after point B, the water pump was turned on.
This is because the desired set temperature range was
95±3°C and a temperature of 92.095±3°C falls within the
acceptable operation range. The water conveyance system
was programmed to switch on the water pump only when
temperatures reached the desired set temperature because
otherwise, the system would pump water at a lower
temperature hence delivering product at a moisture content
lower than the required leading to product loss from
semidried product. The boiler temperature in FIGURE 14.
indicates a drop from 95.75°C to 63.81°C seen at point C
which again occurred at point E. The sudden drop in the
boiler temperature can be attributed to a temporary
malfunction in the DS18B20 temperature sensor
communication with led to this anomaly. However, the two
anomalies never affected the operation of the system since
the pump continued to circulate water throughout the dryer.
This response shows a high robustness of the automation
system to sudden changes which increase its stability. The
minimum and maximum temperatures recorded by the
system were 31.19°C and 98.06°C respectively.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
817
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 14: Relationship between boiler temperature variation and water pump state
Results for performance testing of the synchronized
dryer system
The total drying time of the mango slices was 90 minutes
from beginning to end. The automation system was able to
start belt movement when the water temperature in the
tray reached the desired 85°C after product loading as
shown in FIGURE 15
Throughout the drying process, the belt movement was
smooth with minimal water leakages to the top of the belt
caused by minor perforations in the belt that occurred
during assembly.
FIGURE 15: (A) Mango slices after cutting (B) Loading the dryer with mango slices
From the results, a graph in FIGURE 16 was extracted to
evaluate the relationship between surface temperature,
relative humidity and ambient temperature with drying
time. From the graph, it was observed that in the first 20
minutes of the drying, the surface temperature, ambient
temperature and relative humidity all increased steadily.
This can be attributed to increase in water temperature in
the drying chamber during that time. At the beginning of the
drying process, the relive humidity was 31.4% at point
marked A with which reduced to 18.6% at the end of the
drying process after 90 minutes at point C. This inverse
proportionality relationship of relative humidity with
drying time implies that a lower relative humidity is
required within a drying chamber to have higher drying
rates. Between point A and B, the relative humidity is seen
to increase from 31.4% to 33.5% and this is due to the
placement of the moist fresh mango slices on the dryer belt.
The product surface temperature is seen to increase from
52.7°C at point E to 60.6°C at point F where the graph
descends. The fall in the graph after drying for 50 minutes
at point F can be attributed to the action of the suction
pump inside the drying chamber that was switched on
intermittently. After point F, the surface temperature is
seen to rise and then fall towards the final temperature of
54.2°C at point G. The fall in the graph towards point G is
due to the presence of a cooling section at the end of the
conveyor belt that causes a temperature drop in the water
hence a lower product surface temperature. The ambient
temperature in the drying chamber raises from point H at
44.2°C to I on where then it is seen to drop to point D. The
fall in the dryer ambient temperature could have
happened probably due to operation of the suction pump
on the hood of the drying chamber.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
818
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 16: Relationship between surface temperature, ambient temperature and relative humidity with time
• Moisture content analysis for the Mango slices
Calculation of the moisture content of the slices revealed a
slight variability in the moisture content values for the
slices that were sampled. This variability can be due to the
sizes and shapes of the samples picked form the mango
slices for moisture content calculation hence the sampling
method was a success. TABLE 1 summarizes the moisture
content individual values and the averages there after.
From the table, the graph in FIGURE 18. was extracted to
analyse the relationship between relative humidity inside
the drying chamber and moisture content with drying time
of the RWD. A graph that represents the temperature
differences along the belt during drying was also
superimposed on the same, to understand the temperature
range variations with changes in moisture content along
the belt. After 30 minutes of drying, the highest
temperature difference was recorded as 47.6°C between
the entry and exit of the dryer which significantly reduced
to 27.1°C at the end of the drying process. It can be noticed
that the trend of the temperature range generally reduced
with reduction in product moisture content as marked
from point 1 to 6. The overall reduction in the temperature
difference along the length of the dryer with time is
probably due to the fact that as drying continued, moisture
in the products reduced which reduced the demand for
heat energy from the hot water in the tray during drying
hence increasing temperatures.
TABLE 1: Changes in moisture content with drying time for the mango samples
MC (%)
MC (%)
MC (%)
Sample 1
Sample 2
Sample 3
Average Moisture
content [%]
0
82.9871
71.0081
79.7011
77.90
10
82.5787
70.3298
79.4586
77.46
20
76.2082
58.6317
66.7268
67.19
30
41.5734
55.0379
54.3112
50.31
40
43.4676
54.9236
51.6354
50.01
50
40.8879
50.603
40.877
44.12
60
39.8376
42.0081
39.9709
40.61
70
28.7259
38.3708
30.0519
32.38
80
21.612
19.1571
29.3845
23.38
90
20.1793
18.1305
28.7467
22.35
Drying time (Hr)
Source: Field data
From the graph, it can be seen that the product moisture
content reduces steadily from 77.9% at point 1 to a final
moisture content of 22.4% at point 6. The direct
proportionality between product average moisture
content and RH implies that the relative humidity has a
positive impact on the drying rate of a product. In the first
10 minutes of the drying process, there is a very slight
change in moisture content from 77.9% to 77.5% which
can be attributed to the fact that water temperatures were
still building within the drying bed. Within the last 10
minutes of dying, it is observed that the moisture content
curve almost flattens out towards point 6. This can be
attributed to the presence of a cooling section at the end of
the dryer that reduces the temperature of water hence
reducing the rate of product drying.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
819
International Journal of Scientific Advances
ISSN: 2708-7972
FIGURE 18: Relative humidity changes with moisture content and drying time
Temperature profile along the conveyor belt
The water temperature changes along the conveyor belt
were studied to understand temperature distribution at
different points of the belt with time. The system was able
to switch on the conveyor motor when the temperatures in
the drying bed reached the desired 85°C and this was 22
minutes from the time food was placed on the belt. This
was possible because the water temperature was then
within the acceptable programmed desirable ranges of
85°C ±3. At this point, the water pump was turned on to
start pumping hot water throughout the drying bed. The
recorded temperature changes along the drying bed were
tabulated and the graph in FIGURE 19. extracted.
According to the graph, it is seen that the minimum water
temperature recorded in the system was 31.1 °C at point E
and this was drying after 20 minutes which was 6 m from
the food application side of the dryer. This is because the
water temperature had not picked up enough heat to
distribute it throughout the drying bed. It can be noted that
the average temperature distribution along the conveyor
belt was such that temperatures kept falling at every
recorded drying time from the food application side at
point A towards the product removal side of the dryer at
point C. This is possibly because hot water was entering
from one end of the tray and leaving through an exit pipe
which established a constant temperature gradient along
the drying bed.
FIGURE 19: Temperature changes along the belt during drying
The temperature variations can possibly be reduced by
placing water heater at uninform intervals along the belt
which spreads heat rather than having only one entry of
hot water on the drying bed. Lagging the hood and the
drying bed could also probably reduce the temperature
variations since it reduced heat losses during drying. The
maximum temperature attained by the system was 89.9°C
after 50 minutes of drying marked by point B on the graph
which falls outside the desired temperature range of 85°C
±3. The system was able to reach this temperature
probably because of the significant distance between the
boiler and the entry of the drying bed. This factored in a
significant lag between the time when the upper value of
the desired temperature range was reached and execution
of the command to stop rising temperatures in the boiler.
This lag could possibly be reduced by eliminating the
stand-alone water boiler and inserting heaters directly
within the drying bed. The designed scraper was not able
to easily remove the dried product from the belt after
drying since it was too stuck onto the belt after cooling
hence hand scrapers were used. After several experiments
with the hybrid RWD, it was noticed that it is easier to
remove the product from the belt before cooling hence the
cooling section on the RWD can be eliminated from the
design. The dried mango slices were weighed before and
after drying and results were 775g and 100g respectively.
After drying, the mango slices were able to retain their
original aroma and color which attests to the ability of the
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
820
International Journal of Scientific Advances
dryer to retain the original quality attributes of a product
after drying. The total drying time of mango slices at 3mm
was 90 minutes to attain a final moisture content of 22.4%.
This time can be significantly reduced by reducing the heat
losses in the dryer through design modification.
CONCLUSIONS
From the product performance test done at 85°C, the
conveyor belt of the food conveyance mechanism was able
to move food only when the temperatures reached the
desired temperature range of 85±3°C. This ensured that
food was delivered at the required moisture content which
in turn reduces product loss in the dryer. The lowest
recorded temperature was 31.1°C and the highest
temperature was 89.9°C after 50 minutes. The system
showed results outside the desired range probably
because of the time lag between the switching off the
heater and temperatures stabilizing in the drying bed. This
lag can possibly be reduced by eliminating the boiler and
relocating the heater in the drying bed. A maximum
temperature difference of 47.6°C was recorded along the
drying bed between the beginning and end of the tray
which reduced with reduction in product moisture
content. The reduction in the temperature difference is
possibly due to the reduction in demand of heat energy
from the hot water hence increase in water temperatures
as drying continued. The dried mango slices were
successfully removed from the belt after drying but this
could probably be made easier by eliminating the cooling
section of the dryer. The overall cost and selection of
design materials would promote up scaling and mass
production of the system and hence, it can be installed and
substituted for the manually operated drying system.
Arduino programing software and hardware was found to
work effectively in automating the dryer without
overheating and glitching hence is recommended for
further automation designs of the dryer.
CONFLICTS OF INTEREST
The authors declare that there is no conflict of interest.
ACKNOWLEDGEMENT
This research was funded by the Bioresources Innovations
Network for Eastern Africa Development Programme
(BioInnovate Africa) - Grant#: BA/C1/2017-01_ MAK
ISSN: 2708-7972
REFERENCES
[1] Delta Electronics. (2008). VFD EL USER MANUAL,
1.02(August).
[2] Deshmukh, A., Nagane, M., & Awatade, V. (2013).
Design & Development of Automatic Conveyor
Controlling System for Sorting of Component on
Color basis. Nternational Journal of Science and
Research (IJSR). Retrieved from
http://docplayer.net/52886802-Designdevelopment-of-automatic-conveyor-controllingsystem-for-sorting-of-component-on-colorbasis.html
[3] Groover, M. (2008). Automation, production systems,
and computer integrated manufacturing.
[4] Mujumdar, A. S. (2007). An overview of innovation in
industrial drying: Current status and R&D needs.
Transport in Porous Media, 66(1–2), 3–18.
https://doi.org/10.1007/s11242-006-9018-y
[5] Naphade, A. K., & Deshmukh, P. M. J. (2017). “
Improved Automated Conveyor With Auto
Separation System For Oil Packaging Industry .”
International Research Journal of Engineering and
Technology(IRJET), 4(5), 1245–1251. Retrieved from
https://www.irjet.net/archives/V4/i5/IRJETV4I5243.pdf
[6] Nurus, S. Yogesh, B. (2017). Research Papers
Microcontroller Based Automation of Conveyor Belt,
5(1), 11–16.
[7] Onwude, D. I., Hashim, N., Janius, R., Abdan, K., Chen,
G., & Oladejo, A. O. (2017). Non-thermal hybrid
drying of fruits and vegetables: A review of current
technologies. Innovative Food Science and Emerging
Technologies, 43(November), 223–238.
https://doi.org/10.1016/j.ifset.2017.08.010
[8] Raso, J., & Barbosa-Cánovas, G. V. (2003). Nonthermal
Preservation of Foods Using Combined Processing
Techniques. Critical Reviews in Food Science and
Nutrition, 43(3), 265–285.
https://doi.org/10.1080/10408690390826527
[9] Shirinbakhsh, M. (2017). Design and optimization of
conveyer-belt dryer for biomass, 5(June), 85–94.
Available Online at www.ijscia.com | Volume 2 | Issue 5 | Sep-Oct 2021
821