ARTICLE IN PRESS

BIOSYSTEMS ENGINEERING 99 (2008) 604 – 607

Available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/issn/15375110

Research Note: PH—Postharvest Technology

Drying of copra in a forced convection solar drier

M. Mohanraj, P. Chandrasekar
Mechanical Engineering Department, Dr. Mahalingam College of Engineering and Technology, Pollachi-642003, India

ar t ic l e i n f o A forced convection solar drier was designed, fabricated and tested for the drying copra
under Indian climatic conditions. Drying copra in the drier reduced its moisture content
Article history: from about 51.8% to 7.8% and 9.7% in 82 h for trays at the bottom and top, respectively. The
Received 22 June 2007 copra obtained was graded as 76% milling grade copra (MCG1), 18% (MCG2) and 6% (MCG3)
Received in revised form according to Bureau of Indian standards (BIS: 6220-1971). The thermal efficiency of the solar
13 December 2007 drier was estimated to be about 24%.
Accepted 19 December 2007 & 2008 IAgrE. Published by Elsevier Ltd. All rights reserved.
Available online 12 February 2008

1. Introduction low-quality coconut oil. The oil extracted from poor-quality

copra also requires additional refinement to meet interna-
India ranks as the third largest coconut-producing country tional standards.
in the world. It annually produces 14.37 billion nuts (Singh & Several experimental and theoretical studies have been
Remany, 2002). Copra is one of the major traditional reported on the development of various types of solar driers
products processed from coconuts. Fresh coconut contains for drying agricultural products (e.g. Kadam & Samuel, 2006;
a moisture content of about 52% (wet basis), which should Shanmugam & Natarajan, 2006; Ivanova and Andonov, 2001).
be reduced down to about 7% by drying in order to The main objective of the present work was to study the
concentrate the oil content. On an average, 5–7 coconuts are drying characteristics and quality of copra produced in a
required to produce 1 kg of copra, although this depends on forced convection solar drier so that it could be recommended
the source. to farmers for high-quality copra production.
The traditional methods followed in India are sun and kiln
drying. They produce poor-quality copra and are time
consuming. With kiln drying, smoke is in direct contact with 2. Material and methods
the coconut cups. As a result, high-quality copra is not
produced and smoke deposits may form polycyclic aromatic The experiments were carried out at the coconut farm in
hydrocarbons in the copra (Thiruchelvam et al., 2007). Pollachi, India, from January to April 2007.
Sun drying takes about 7 days and if the weather is rainy
the copra produced will be contaminated with fungi which 2.1. Forced convection solar drier
produce a grey rancid product. Furthermore sun drying
requires more space, is labour intensive and there can be A schematic diagram of a forced convection solar drier is
deteriorations in quality from deposits of dirt and dust. shown in Fig. 1(a) and a cross-sectional view of a solar air
Also, microorganisms can increase the acid content, cause heater in Fig. 1(b). The solar drier consisted of a flat plate solar
rancidity and reduce the amount of extractable oil resulting in air heater of area 2 m2 (2 m  1 m) connected to a drying

Corresponding author. Tel.: +91 9486411896; fax: +91 4259 236070

E-mail address: [email protected] (M. Mohanraj).
1537-5110/$ - see front matter & 2008 IAgrE. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.biosystemseng.2007.12.004
 ARTICLE IN PRESS
BIOSYSTEMS ENGINEERING 99 (2008) 604– 607 605

Humid air
Exit 0.4m
Ambient Dry and wet Glass wool insulation
bulb temperature 0.6m

Loading
Copra
V Velocity of air at tray entry Door
Wet bulb temperature Drying chamber
1.5m
Dry bulb temperature
Trays
Solar air heater V
Absorber plate A
Sand
Glass cover 1m
Orificemeter A1
Control valve
Blower Insulation
Manometer

25
5

50 100
1000
All dimensions are in mm

Fig. 1 – (a) Schematic view of the solar drier used for copra drying. (b) Sectional view A–A1 of a solar air heater.

chamber. The solar air heater had a 2 mm thick copper (latitude of 10.391N, longitude of 77.031E), where the experi-
absorber plate coated with black paint to absorb the incident ment was conducted, had about 11 h 30 min of daylight, with
solar radiation. The absorber plate was placed directly behind typically about 8 h per day of sunshine available for drying.
the transparent cover (glass) with a layer of air separating Four calibrated thermocouples (Pt 100) with70.25 1C accu-
it from the cover. The air to be heated passed between the racy were fixed at different locations (as shown in Fig. 1) of
transparent cover (glass) and the absorber plate. To reduce the the solar drier to measure the temperature of drying air
losses from topside of the absorber plate and to increase through a digital scanner having a 0.1 1C resolution connected
the temperature of air using the greenhouse effect, a 5 mm with a rotary selector switch. Power input to the blower was
thick glass cover was placed over the absorber. The gap measured with an energy meter having 70.4% accuracy.
between the glass and the absorber surface was maintained A U-tube manometer was fixed in the path of the air circuit to
at 25 mm for air circulation. One side of the collector was measure the velocity of air entering the drier. The velocity of
connected to a 0.75 kW (1 HP) centrifugal fan with an airflow air at the inlet of the tray was measured with the help of a
rate up to 300 m3 h1 and the other side with a drying vane-type anemometer having70.01 m/s accuracy. Solar
chamber. A divergent section was provided at the entry of intensity was measured using a solar intensity meter having
the solar air heater to provide uniform air circulation over the an accuracy of about 72%. A digital electronic balance of 1 kg
absorber surface. The 100 mm gap between the absorber and capacity having an accuracy of 70.001 g was used to weigh
the insulation was filled with sand to store heat. The drying the samples. The humidity of the ambient air was measured
chamber was made up of a 2 mm thick mild steel sheet with a using standard non-aspirated wet and dry bulb mercury
width, depth and height of 1 m  1 m  1.5 m, respectively. The thermometers with sensitivities of 0.5 1C.
drying chamber was insulated with 10 mm thick glass wool.
The solar air heater was tilted to an angle about 251 with 2.2. Experimental procedure
respect to the horizontal, which is considered to be an
optimum angle for year-round performance of the system at Broken coconuts were loaded over the trays of a drier
Pollachi (Shariah et al., 2002). The system was oriented to face chamber. For hot air circulation over the products to be dried,
the south to maximise the incident solar radiation on the the wire mesh trays had a porosity of about 90% . Then the
solar collector. On the basis of measurements, Pollachi fan was switched on and the air velocity adjusted to an
 ARTICLE IN PRESS
606 BIOSYSTEMS ENGINEERING 99 (2008) 604 – 607

optimum of 1.2 m/s at the tray inlet. The velocity of the air at 1200 100
Solar intensity
the tray was adjusted by using a control valve. During the 90
Ambient relative humudity
experiments, temperatures at various locations in the solar 1000 80

Solar radiation (W m-2)

Relative humudity (%)

collector and the drier chamber, ambient dry and wet bulb
70
temperatures were measured at hourly intervals. The relative 800
60
humidity of air was calculated from measured wet and dry
bulb temperatures using a psychometric chart. After the 600 50
moisture content was reduced to 40%, the copra kernels 40
400 30
were scooped from the shells and dried further without
shells. Moisture contents were determined by using Eq. (1). To 20
200
assure the quality of copra obtained, three experiments were 10
carried out. At the end of each experiment, the dried copra 0
0
was graded according to Bureau of Indian standards (BIS: 0 6 12 18 24 30 36 42 48 54 60 66 72 78
6220-1971). The average grade value is presented. Experi- Drying time (h)
ments were only conducted during daylight hours.
Fig. 2 – Variation of solar intensity and ambient relative
2.3. Data analysis humidity.

The quantity of moisture content in a material can be 70

represented on a wet basis and expressed in percentage. About
60
10 g samples were chopped from randomly selected five cups
and kept in a convective electrical oven, which was maintained Temperature (°C) 50
at 10571 1C for 4 h. The initial (Wi) and final (Wf) mass of the
40
samples was recorded with the help of an electronic balance.
The moisture content (Mwb) on a wet basis was calculated by 30
ðWi  Wf Þ 20
Mwb ¼  100. (1) Tda
Wi
Tamb
10
The instantaneous thermal efficiency of the solar air heater
was estimated by using Eq. (2) according to (Kadam & Samuel, 0
2006) 0 8 16 24 32 40 48 56 64 72 80
cp ma ðTo  Ta Þ
Drying time (h)
Zth ¼  100 (2)
AS I Fig. 3 – Variation of ambient and drying air temperature as a
where cp is the specific heat of air in J kg1 K, ma is the mass function of drying time.
flow rate of air in kg s1; Ta is the ambient temperature and To is
the outlet air temperature. A is the surface area of the solar
collector in m2 and I is the solar intensity in W m2. 60
Moisture content (wet basis) (%)

50
Top
3. Results and discussion Bottom
40
The variation of solar radiation and ambient relative humi-
30
dity during experimentation is shown in Fig. 2. A maximum
solar intensity of 932 W m2 was observed. The ambient 20
relative humidity varied between 55% and 72% with an
average of about 68% . Temperature variations of the drying 10
air and ambient are shown in Fig. 3. The average drying air
temperature recorded at the inlet of the drier was 43 1C. The 0
maximum drying air temperature recorded during peak 0 8 16 24 32 40 48 56 64 72 80
sunshine hours was 63 1C. The average temperature reduced Drying time (h)
to 31 1C outside the hours of sunshine and during the night.
Fig. 4 – Variation of moisture content as a function of drying
At the outlet of the drying chamber, a high relative humidity
time.
of about 90% was recorded during the initial stages of drying
but this gradually reduced to about 34% at the end of drying.
The variation of moisture content (wet basis) with drying moisture reduction during the first day was observed because
time is shown in Fig. 4. The average moisture content of the of the evaporation of free moisture migrating from the
coconut was reduced from about 51.8% to 7.8% and 9.7% in the outer surface layers. This then reduced due to the internal
bottom and the top tray, respectively, after 82 h. The moisture migration of moisture from inner layers to the surface
reduction during the first and the second day of drying was producing a uniform dehydration of wet kernel. Typically,
found to be about 33% and 20%, respectively. The higher 10 h a day of drying occurred during sunshine. Without
 ARTICLE IN PRESS
BIOSYSTEMS ENGINEERING 99 (2008) 604– 607 607

sunshine desorption of moisture took place and the moisture forced convection solar drier is more suitable for produ-
content increased by about 0.5–1.5%. The reduction in the cing high-quality copra for small holders. About 75% of
moisture content of copra at the bottom tray was about 5–8% high-quality copra (MCG1) could be produced. The average
higher than that of the top tray. A high drying rate at a rate of thermal efficiency of the solar air heater was estimated to be
4.2 g of water/g of dry matter was observed during the initial about 24%.
stage of drying. The drying rate decreased with an increase in
the drying time. Drying occurs in the falling rate period with a R E F E R E N C E S
steep fall in the moisture content in the initial stages of drying
which becomes very slow in the later stages. The drying rate of
copra in the solar dryer was high compared to sun drying due Shariah A; Al-Akhras M -A; Al-Omari I A (2002). Optimizing
to its high heat and mass transfer coefficients. the tilt angle of solar collectors. Renewable Energy, 26,
About 50 kg of moisture was removed from 300 nuts to 587–598
Ivanova D; Andonov K (2001). Analytical and experimental study
obtain about 60 kg of copra. The initial weight of 300 nuts with
of combined fruit and vegetable dryer. Energy Conversion and
shells was measured to be about 160 kg. The thermal
Management, 42, 975–983
efficiency of the solar air heater was estimated to be about Kadam D M; Samuel D V K (2006). Convective flat plate solar heat
24% by using Eq. (2). collector for cauliflower drying. Biosystems Engineering, 93,
About 60 kg of copra was produced from 300 nuts. The copra 189–198
obtained was graded as 76% MCG1, 18% MCG2 and 6% MCG3. Thiruchelvam T; Nimal D A D; Upali S (2007). Comparison of
Based on the grading of copra, it could be concluded that quality and yield of copra processed in CRI improved kiln
drying and sun drying. Journal of Food Engineering, 78,
more than 75% of high-quality MCG1 could be produced in the
1446–1451
solar drier. Unlike with kiln and sun drying, the copra Shanmugam V; Natarajan E (2006). Experimental investigation of
obtained was free from smoke, dust, bird and rodent damage. forced convection and desiccant integrated solar dryer.
Renewable energy, 31, 1239–1251
Singh H P; Remany G (2002). Approaches for increasing the farm
4. Conclusion income through product diversification and product utiliza-
tion. In: Sustainable Products and Utilization of Coconut
A forced convection solar drier was designed, fabricated (Singh HP; Mathew MT, eds), pp 1–11. Coconut Development
and tested for drying copra. It can be concluded that a Board, Kochi