AL SCIE

UR NC
AT
N

E
ED AND

FO
UNDATI O
Journal of Applied and Natural Science 7 (2) : 939 - 948 (2015) JANS
PLI

ANSF
AP

2008

Development and evaluation of multi millet thresher

K. P. Singh*, Rahul R. Poddar, K. N. Agrawal, Smrutilipi Hota and Mukesh K. Singh
Agricultural Mechanization Division, Central Institute of Agricultural Engineering, Bhopal-462038, (Madhya Pradesh),
INDIA
*Corresponding author. E-mail: [email protected]
Received: March 14, 2015; Revised received: October 21, 2015; Accepted: November 18, 2015

Abstract: In tribal areas of India, traditional methods of threshing of minor millets like little millet (Panicum suma-
trense), M1, kodo millet (Paspalum scrobiculatum), M2, foxtail millet (Setaria italica), M3, proso millet (P. miliaceum),
M4, barnyard millet (Echinochloa frumantacea), M5, finger millet (Eleusine coracana), M6 is done of beating by sticks
or treading out the crop panicle under the feet of oxen. This operation is most time consuming, labour intensive,
drudgery prone, uneconomical, lower output and obtain low quality products. A thresher for these millet crops was
developed and optimization of the operating parameters with little millet was done by using Response surface meth-
odology (RSM). The optimized parameters were 7.79% (d.b) moisture content, 105 kgh-1 feed rate, 625 rpm cylinder
speed, 5 mm threshing sieve size which gave maximum threshing efficiency of 95.13% and cleaning efficiency of
94.12%. After optimization of parameters the thresher was tested for threshing of all the six minor millets with proper
adjustments of sieve. Threshing capacity of M1, M2, M3, M4, M5 and M6 were obtained as 89, 137, 140, 91, 88 and 99
kg/h, respectively with more than 96% threshing efficiency and less than 2% broken grain.
Keywords: Cleaning efficiency, Multi millet thresher, Shear and impact cutting, Threshing efficiency

INTRODUCTION threshing operations are most time consuming, energy

intensive (19.9 kJ/min), labour intensive, drudgery
Millets are very important food crop of tribal people prone and uneconomical. The mechanized threshing of
and believed to be first domesticated cereal crop which millets can reduce the drudgery of farmers/labours,
is mostly grown in tribal and hilly areas of India and improve the quality of product. With existing socio
many Asian and African nations (Gbabo et al., 2013). economic condition of millet growing tribal farmers,
These crops are grown in rain fed areas and tempera- the large capacity threshers are inappropriate and even
ture more than 20°C where other crops yield are very the small size thresher with large scale sophistication
poor and are less prone to disease and pests. In India, are difficult to be adopted (Singh et al., 2002). World-
millets are major staple food in the state of Uttara- wide number of studies have been done for threshing
khand, Chhattisgarh, Tamil Nadu, Odisha and Karna- of various crops but a very few studies has been re-
taka where these are grown widely with yield as high ported on millet threshing. Therefore, development of
as 3 t/ha. Maximum production (thousand tonnes), thresher for all millet crops was found necessary which
productivity (kg/ha) and area of production (thousand can do threshing of all millets.
ha) of different millets in India are of Finger millet Physical properties of crop are very important for the
(1964.9, 1179, 1641.6), followed by Barnyard millet design and development of machine. Singh et al.
(180.1, 863, 208.6), Kodo millet (146.3, 366, 399.4), ( 2010) studied different physical properties of barn-
Little millet (113.2, 364, 310.9), Foxtail millet (62.9, yard millet. Baryeh (2002) evaluated different physical
607, 103.7) and Proso millet (26.4, 556, 47.5). Millets properties of millets and expressed as function of
are very nutritious and important crop for balanced moisture content between 5−22.5%. (Singh et al.,
diet, rich in vitamins, protein, carbohydrate, minerals, 2003) developed a thresher at Vivekananda institute of
fibers, iron, amino acid, phosphorus, magnesium, po- Hill Agriculture (ICAR), Almora, Uttarakhand, India
tassium and also good source of energy. The epidemi- and was modified at IIT, Kharagpur. It was observed
ological evidences indicate that person on millet based that threshing of millet is better in case of combination
diets have good resistance for degenerative diseases of impact and shear on the crop. Therefore the machine
such as heart disease, diabetes, hypertension etc for threshing of all six minor millets was developed on
(Anonymous, 2001). the principle of combination of shear and impact at
Traditionally in tribal and hilly areas, threshing of mil- Central Institute of Agricultural Engineering, Bhopal,
let crop is done either beating by sticks or by treading India.
out the crop panicle under the feet of oxen. These Many researchers have worked on the optimization of
ISSN : 0974-9411 (Print), 2231-5209 (Online) All Rights Reserved © Applied and Natural Science Foundation www.ansfoundation.org
940 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015)

process parameters like milling, threshing (Singh et al., vas strips and three rows of cutting knives places alter-
2004; Tiwari et al., 2007; Singh et al., 2008). Ajav and nately as some of the millet crop requires cutting ac-
Adejumo (2005) studied the performance evaluation tion and some requires shear for complete threshing.
by taking moisture content, cylinder speed and feed The knives arrows provides impact cutting of crop
rate as independent parameter to obtain the maximum stem during threshing and the canvas strip rows gives
threshing efficiency. Kushwaha et al. (2005) devel- gentle abrasion and shear on the grain for removing the
oped an okra seed extractor and evaluated the effect of grains from the glumes. The threshing chamber is fit-
drum speed and moisture content on extracting effi- ted with a sliding sieve which is allow repetitive im-
ciency. Singh et al. (2008) developed a pedal operated pact and shear to complete detach of glumes from the
paddy thresher and optimized the independent parame- grains which helps in complete threshing of millets.
ter like drum speed for getting highest threshing effi- Ergonomic consideration in the design of the
ciency. Similarly Singh et al., (2010) optimized the thresher: In general these crops threshed inside the
value of drum speed for threshing of finger millet. boundary of the house. The ergonomic and safety is
Response surface methodology (RSM) is defined as very important especially for the use of tribal women
the statistical method that uses quantitative data from worker. Ergonomic considerations were used in the
an appropriate experimental design to determine and design of thresher for safety of the worker. The length
simultaneously solve multivariate equations. The main of the feeding chute was kept 900 mm as per IS:
advantage of RSM is that it reduces the number of 9020−2002. Grip handles of threshing sieves were
experiments needed to evaluate multiple parameters made as per inner grip diameter for better comfort of
and their interactions. It was used successfully by the worker. Shaker assembly for cleaning system was
many scientists for optimization of different parame- provided with packing for reduction of vibration and
ters for different operations (Goyal et al., 2008; Singh noise. Rubber transportation wheels instead of cast
et al., 2008; Ushakumari et al., 2007; Nath and Chat- iron wheels were provided for easy transportation and
topadhyay, 2007). The present study was undertaken to for absorption of vibration during operation. The ma-
use RSM to optimize the operational parameters: mois- chine is attached with safety guards over power trans-
ture content(MC), feed rate(FR), drum speed(DS), mission system. A flapper was fitted in the feeding
threshing sieve size(TSS) to maximize the threshing chute to arrest the dust which may create health prob-
and cleaning efficiencies of the thresher. lems of the worker.
Evaluation of multi millet thresher: The multi millet
MATERIALS AND METHODS thresher prototype was evaluated after development.
Raw material: Different millets like little, kodo, For the evaluation little millet crop was taken after
proso, foxtail, barnyard and finger millet of local varie- harvesting. The crop was taken as the whole crop for
ties were collected from small village Patalkot/ threshing. But after the optimization the final thresher
Dhindhori tribal areas of MP. Physical properties of all was tested with all six minor millets by taking only the
millets (Table 1) were observed to suitable for the ear heads of kodo millet, foxtail millet, barnyard millet
proper threshing of millet. The moisture content of the and finger millet except little millet and proso millet as
crops was kept 5 to 11% for performance analysis and recommended by the tribal farmers. The feed rate was
digital moisture meter was used for this purpose. controlled manually by the help of a worker and main-
Design and development of the thresher: A multi tained from 75 to 120 kgh-1. They had to feed the
millet thresher was designed and developed for thresh- whole crop as per requirement. A two hp, single phase
ing of the millets based on the different properties of electric motor was used as power source and the power
the minor millets. Developed machine works on the is transmitted to the threshing drum, aspirator by the
principle of impact and shear force acting on the ear help of belt drive. For variation of drum speed from
head of the crop for the purpose of threshing of millets. 500 to 1000 rev/min different size of pullies were used
The threshing drum was fitted with three rows of can- according to the requirement. The air flow rate was

Table 1. Specifications of the developed CIAE multi millet thresher.

Length 2027 mm
Width 1048 mm
Height 1200 mm
Weight of the machine 180 kg
Threshing chute length 900 mm
Cost of the machine Rs 45,000/-
Power requirement 2 hp, single phase electric motor
Machine noise level 84 db
Threshing capacity 80-150 kg/h (depending on millet comodity)
Dehulling capacity 20-80 kg/h (depending on millet comodity)
Pearling capacity of finger millet 200-250 kg/h
 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015) 941

Table 2. Coded values and corresponding real values used in experimentation.

Independent variable Coded value –α (–1.414) –1 0 +1 + α (+1.414)
Moisture content (Mc), % db Actual value 3 5 7 9 11
Feed rate (Fr), kg/h 60 75 90 105 120
Drum speed (Ds), rpm 500 625 750 875 1000
Threshing sieve size (Tss), mm 3 5 7 9 11
Table 3. Experimental design for thrshing of little millet using CCRD with four independent variables.

Independent variables Dependent variable

Moisture content Feed Rate Drum Speed Threshing Sieve Threshing Cleaning
Run (Mc), % (db) (Fr), kg/h (Ds), rpm Size (Tss), mm efficiency efficiency
(Te), % (Ce), %
1 5 (-1) 75(-1) 875(+1) 5(-1) 97 96
2 7 (0) 90(0) 750(0) 11(+α) 92 93
3 9 (+1) 105(+1) 625(-1) 9(+1) 85 87
4 9(+1) 105(+1) 875(+1) 9(+1) 90 89
5 9(+1) 75(-1) 875(+1) 5(-1) 94 94
6 5(-1) 105(+1) 875(+1) 9(+1) 93 91
7 9(+1) 75(-1) 875(+1) 9(+1) 91 91
8 9(+1) 105(+1) 875(+1) 5(-1) 96 93
9 7(0) 90(0) 1000(+α) 7(0) 99 97
10 5(-1) 75(-1) 625(-1) 9(+1) 91 88
11 7(0) 90(0) 750(0) 7(0) 90 96
12 9(+1) 105(+1) 625(-1) 5(-1) 95 90
13 7(0) 60(–α) 750(0) 7(0) 96 93
14 9(+1) 75(-1) 625(-1) 5(-1) 90 87
15 11(+α) 90(0) 750(0) 7(0) 84 86
16 5(-1) 105(+1) 875(+1) 5(-1) 95 94
17 9(+1) 75(-1) 625(-1) 9(+1) 88 92
18 7(0) 90(0) 750(0) 7(0) 89 94
19 7(0) 120(+α) 750(0) 7(0) 94 90
20 5(-1) 75(-1) 875(+1) 9(+1) 96 95
21 7(0) 90(0) 750(0) 7(0) 94 93
22 3(–α) 90(0) 750(0) 7(0) 90 89
23 7(0) 90(0) 750(0) 3(–α) 93 96
24 7(0) 90(0) 750(0) 7(0) 92 93
25 7(0) 90(0) 500(–α) 7(0) 96 89
26 7(0) 90(0) 750(0) 7(0) 92 95
27 5(-1) 105(+1) 625(-1) 5(-1) 95 90
28 5(-1) 75(-1) 625(-1) 5(-1) 93 90
29 7(0) 90(0) 750(0) 7(0) 92 90
30 5(-1) 105(+1) 625(-1) 9(+1) 93 89

maintained below the terminal velocity of the grains

and above the terminal velocity of chaff. The speed of ….…………. (1)
aspirator was maintained by the use of belt drive.
The moisture content was varied by adding of water to The evaluation was done by taking the required sample
the sample. The samples were prepared by spraying of 10 kg each time and the data were collected. Some
the desired amount of distilled water to the samples, formulas used for calculation of threshing efficiency,
thoroughly mixed, sealed in separate polyethylene cleaning efficiency were,
bags and then kept for a week for uniform distribution Threshing efficiency = threshed grains received from
of moisture throughout the samples. Before each ex- all the outlets with respect to total grain input ex-
periment required amount of samples were taken out pressed as percentage by mass.
and allowed to warm up to the room temperature TE=100–percentage of unthreshed grains ..............(2)
(Goyal et al., 2008; Singh et al., 2010). The amount of Percentage of unthreshed grains
water required for desired moisture content was calcu-
lated by the following equation (Karababa, 2006; Al-
tuntas and Yildiz, 2007; Singh et al., 2010). ............. (3)
942 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015)
Table 4. Analysis of variance for threshing efficiency.

Sum of Mean
Source Squares df Square F Value p- value > F
Model 283.33 14 20.23 6.57 0.0004 Significant
Mc 54.00 1 54.00 17.54 0.0008
Fr 0.16 1 0.16 0.05 0.8191NS
Ds 32.66 1 32.66 10.61 0.0053
Tss 37.50 1 37.50 12.18 0.0033
Mc×Fr 1.00 1 1.00 0.32 0.5771NS
Mc×Ds 1.00 1 1.00 0.32 0.5771NS
Mc×Tss 12.25 1 12.25 3.98 0.0645NS
Fr×Ds 6.25 1 6.25 2.03 0.1746NS
Fr×Tss 9.00 1 9.00 2.92 0.1079NS
Ds×Tss 1.00 1 1.00 0.32 0.5771NS
Mc2 38.67 1 38.67 12.56 0.0029
Fr2 18.10 1 18.10 5.88 0.0284
Ds2 56.67 1 56.67 18.41 0.0006
Tss2 0.96 1 0.96 0.31 0.5839NS
Lack of Fit 30.66 10 3.06 0.98 0.5402 not significant
Table 5. Analysis of variance for cleaning efficiency.

Sum of F p-value
Mean Square
Source Squares Df Value Prob > F
Model 223.33 14 15.95 5.79 0.0008 Significant
Mc 10.67 1 10.67 3.87 0.0679NS
Fr 88.17 1 88.17 32.00 < 0.0001
Ds 13.5 1 13.5 4.9 0.0428
Tss 0.00 1 0.00 0.00 1 NS
Mc×Fr 4.00 1 4.00 1.45 0.2469NS
Mc×Ds 0.25 1 0.25 0.091 0.7674NS
Mc×Tss 4.00 1 4.00 1.45 0.2469NS
Fr×Ds 6.25 1 6.25 2.27 0.1528NS
Fr×Tss 6.25 1 6.25 2.27 0.1528NS
Ds×Tss 70.58 1 70.58 25.61 0.0001
Mc2 10.01 1 10.01 3.63 0.076 NS
Fr2 1.44 1 1.44 0.52 0.4808NS
Ds2 0.58 1 0.58 0.21 0.652NS
Lack of Fit 19.83 10 1.98 0.46 0.8606 Not significant

Where, D =Quantity of unthreshed grains collected CCRD for optimization of operational parameters:
from all outlets per unit time The operational parameters were fixed at 5 levels
A =Total grain input per unit time (Table 2) as per CCRD and a total number of 30 ex-
Cleaning efficiency = Clean grain received at main grain periments were carried out (Table 3). Seven repeated
outlet with respect to the total grain mixture received at experiments were conducted at the central points of the
main grain outlet expressed as percentage by mass. coded variables to calculate the error sum of squares
and the lack of fit of the developed regression equation
between the responses and independent variables.
............... (4) The parameters were optimized by using Design expert
Where, M = Quantity of clean grain obtained from the 7.0.0 software, which gave optimum values based on
sample taken at main grain outlet. experimented results.
F = Total quantity of the sample taken at main grain outlet.
Experimental design: Central Composite Rotatable RESULTS AND DISCUSSION
Design (CCRD) (Rastogi et al., 1998; Singh et al., Threshing efficiency: Analysis of variance was con-
2011) was considered as experimental design with four ducted for threshing efficiency and is presented in the
independent parameters moisture content, feed rate, table 4. The ANOVA data shows high F value (6.58)
threshing drum speed and threshing sieve size for optimi- which implies the model to be significant at 0.1% level
zation. The responses were obtained in terms of threshing of significance. The linear effect of moisture content
and cleaning efficiency and optimized by use of RSM. was highly significant on threshing efficiency at 0.1%
 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015) 943

96 96

93 94.25
TE, %

90

TE, %
92.5

87 90.75

84 89

105.00 9.00 9.00 105.00

97.50 8.00 8.00 97.50
90.00 7.00 7.00 90.00
FR, kg/h 82.50 6.00 MC, % (db) TSS, mm 6.00 82.50 FR, kg/h
75.00 5.00 5.00 75.00

(a) (d)

94 99

91.5 96.5

89 TE, % 94
TE, %

86.5 91.5

84 89

9.00 9.00 875.00 105.00

8.00 8.00 812.50 97.50
7.00 7.00 750.00 90.00
TSS, mm 6.00 6.00 MC, % (db) DS, RPM 687.50 82.50 FR, kg/h
5.00 5.00 625.00 75.00

(b) (e)

99 99

95.25 96.5
TE, %

91.5
TE, %

94

87.75 91.5

84 89

875.00 9.00 9.00 875.00

812.50 8.00 8.00 812.50
750.00 7.00 7.00 750.00
DS, RPM 687.50 6.00 MC, % (db) TSS, mm 6.00 687.50 DS, RPM
625.00 5.00 5.00 625.00

(c) (f)
Fig. 1. Response surface contours for threshing efficiency of millets as a function of (a) Feed rate moisture content, (b)
threshing sieve size and moisture content, (c) drum speed and moisture content, (d) threshing sieve size and Feed rate, (e)
drum speed and Feed rate and (f) threshing sieve size and drum speed.

level of significance and effect of drum speed and variables neglecting the high error generating terms
threshing sieve size were significant at 1% level of was presented in the equation 5.
significance. The quadratic term of drum speed is Te = +91.50 -1.50 Mc +1.17 Ds -1.25Tss -1.19
highly significant (p<0.001) and moisture content Mc2+0.81 Fr2+1.44 Ds2………………….(5)
(p<0.01) and feed rate (0.05) had significant effect on Where, Te = Threshing efficiency, %
threshing efficiency. The lack of fit was obtained non Mc = Moisture content, % (d.b)
significant on threshing efficiency. No significant ef- Ds= Drum speed, rev/min
fect was found in case of all the interactions of the Tss= Threshing sieve size, mm
variables. The regression equation obtained for the Fr= Feed Rate, kgh-1
response threshing efficiency with four independent
944 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015)

96 97

93.5 94.25

91.5

CL, %
91
CL, %

88.5 88.75

86 86

105.00 9.00 875.00 9.00

97.50 8.00 812.50 8.00
90.00 7.00 750.00 7.00
FR, kg/h 82.50 6.00 MC, % (db) DS, RPM 687.50 6.00 MC, % (db)
75.00 5.00 625.00 5.00

(a) (d)

97 96

95 93.5

93 91
CL, %
CL, %

91 88.5

89 86

875.00 105.00 9.00 9.00

812.50 97.50 8.00 8.00
750.00 90.00 7.00 7.00
DS, RPM 687.50 82.50 FR, kg/h TSS, mm 6.00 6.00 MC, % (db)
625.00 75.00 5.00 5.00

(b) (e)

96 97

94.5 95
CL, %

93
CL, %

93

91.5 91

90 89

9.00 105.00 9.00 875.00

8.00 97.50 8.00 812.50
7.00 90.00 7.00 750.00
TSS, mm 6.00 82.50 FR, kg/h TSS, mm 6.00 687.50 DS, RPM
5.00 75.00 5.00 625.00

(c) (f)

Fig. 2. Response surface contours for cleaning efficiency of millets as a function of (a) Feed rate and moisture content, (b)
drum speed and feed rate, (c) threshing sieve size and feed rate (d) drum speed and moisture content, (e) threshing sieve size
and moisture content and (f) threshing sieve size and drum speed.

Response surface plots and contours of threshing effi- mental range of feed rates, threshing sieve size and
ciency as function of moisture content, feed rate, drum drum speed (Fig. 1a, 1b, 1c). It can be observed from
speed and threshing sieve size are showed in Fig 1 a, b, Fig. 1a that at a fixed feed rate of 75 kgh-1 threshing
c, d, e and f. Threshing efficiency was found slow de- efficiency was decreased from 92 to 91.5% slowly as
creased with increase of moisture content from 5 to 7% the moisture content was increased up to 7% and de-
and decrease was rapid thereafter within the experi- creased rapidly thereafter up to 90%. Kamble et al.
 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015) 945
Table 6. Comparison of physical and threshing parameters of selected minor millets.
Millets Dia Sphericity Bulk Thousand Thresh- Thresh- Broken Clean-
(mm) (fraction) density grain ing ca- ing effi- grain ing effi-
(kg/m3) weight pacity ciency (%) ciency
(gm) (kg/h) (%) (%)
Panicum suma- 2.26b 0.481a 815.0c 2.08a 88.7a 97.57a 1.70b 97.13a
trense (M1)
Paspalum scrobi- 3.41b 0.818e 747.0a 4.20d 116.7b 99.33b 0.50a 98.96b
culatum (M2)
Setaria italic 2.35b 0.512b 753.7a 2.39b 95.3a 99.40b 1.20b 99.12b
(M3)
Panicummili- 1.95a 0.485a 825.3c 2.32b 95.0a 97.33a 1.60b 96.98a
aceum (M4)
Echinoch- 2.32b 0.544c 895.3d 3.35c 108.3b 96.33a 0.54a 95.89a
loa frumantacea
(M5)
Eleusine cora- 1.63a 0.681d 802.0b 2.35b 102.0a 99.40b 0.00a 99.27b
cana (M6)

Overlay Plot Overlay Plot

1100.00 10.00

Threshing Efficiency: 100

950.00 8.75
Threshing Eff 99.7461 Threshing Efficiency: 95
Moisture content, % (db)

Cleaning Effic 87.5129

Drum speed, RPM

X1 6.01591
X2 7.79

800.00 Cleaning Efficiency: 89 7.50

Cleaning Efficiency: 87
Threshing Efficiency: 95
Threshing Efficiency: 100
Threshing Eff 99.5279
99.5124
Cleaning Effic 88.0521
88.0444
X1 6.06
6.07 Cleaning Efficiency: 87
X2 626.90
650.00 6.25

Threshing Efficiency: 100

500.00 5.00
5.00 6.00 7.00 8.00 9.00 5 6 7 8 9

T hreshing sieve size, mm T hreshing sieve size, mm

(a) (c)

Overlay Plot Fig. 3. Super imposed contours plots of (a) drum speed and
1 0 .0 0 threshing sieve size, (b) moisture content and feed rate and
(c) moisture content and threshing sieve size.

8 .7 5
creased the plasticity of the grain. Bansal and Lohan
Moisture content, % (db)

Threshing Eff 99.1248

Cleaning Effic 88.5744
X1 105.31
Threshing Efficiency:
X2 7.79
95 (2009) obtained higher threshing efficiency at lower
moisture content during threshing of seed crops.
7 .5 0 Threshing Efficiency: 100
Kushwaha et al. (2005) developed an Okra seed ex-
tractor and evaluated at different moisture content
6 .2 5
from which they got the result as extracting efficiency
was decreased with increase of moisture content.
Fulani et al. (2013) resulted that higher threshing
5 .0 0
Cleaning Efficiency: 87 efficiency observed at lower moisture content in case
50 .00 7 5 .0 0 1 0 0 .0 0 1 2 5 .0 0 1 50 .0 0 of cowpea. Increase in threshing efficiency was ob-
served very slowly with increase of feed rate within all
Feed rate, kg/h
experimental range of moisture content, threshing
(b) sieve size and drum speed (Fig. 1a, 1d, 1e). Abo-El-
Naga et al. (2015) found increased threshing efficiency
(2003) conducted the study of pearl millet thresher and with increase of feed rate. In the present study, thresh-
got the reduced threshing efficiency with increase of ing efficiency was found to be increased rapidly with
moisture content because high moisture content in- increased of drum speed with all experimental range of
946 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015)

moisture content, feed rate and threshing sieve size Fr = Feed Rate, kgh-1
(Fig. 1c, 1e, 1f). At a fix moisture content of 5% the Response surface plots and contours of cleaning effi-
threshing efficiency was increased from 88 to 95% as ciency as function of moisture content, feed rate, drum
the drum speed increased from 625 to 875 rev/min. speed and threshing sieve size are showed in Fig 2 a, b,
This may be due to more impact action of the drum per c, d, e and f. Cleaning efficiency was found slowly
unit time on the crop. Kamble et al., 2003 studied the decreased when moisture content increased from 5 to
effect of drum speed on threshing of pearl millet. It 7% and decreased thereafter rapidly within the experi-
was observed that increase in drum speed increased the mental range of feed rates, threshing sieve size and
threshing efficiency in a high range. Bansal and Lohan drum speed (Fig. 2a, 2d, 2e). It can be observed from
( 2009) also found higher threshing efficiency at high Fig. 3a that at a fixed feed rate of 75 kgh-1 the cleaning
drum speed in case of seed crops. Ajav and Adejumo efficiency was decreased slowly as the moisture con-
(2005) studied performance evaluation of an okra tent was increased up to 7% and decreased rapidly
thresher and got the similar result as increase in cylin- thereafter up to 91%. Bansal and Lohan( 2009) re-
der speed increased the threshing efficiency. Singh et ported higher cleaning efficiency at lower moisture
al., 2008 studied the effect of threshing drum speed on content in case of seed crops. Simonyan et al., (2006)
threshing performances of pedal operated VL paddy also reported same result in case of cleaning perform-
thresher and got the same result. Sinha et al. (2007) ance of stationary sorghum thresher. Fulani et al.
studied the effect of drum speed on chickpea seed crop ( 2013) found higher cleaning efficiency at lower
threshing, Fulani et al. (2013) also found similar result moisture content during threshing of cowpea thresher.
that threshing efficiency increased with increase of Cleaning efficiency was observed slowly decreasing
drum speed. Simonyan and Imokheme (2008) studied with increase of feed rate in case of all the experimen-
the effect of drum speed on sorghum threshing and tal range of moisture content, threshing sieve size and
they also got similar observations of increased thresh- drum speed (Fig. 2a, 2b, 2c). This may be due to the
ing efficiency with increase of drum speed as in pre- increase of grain handling due to increase of feed rate.
sent study (Fig. 1c). Threshing efficiency was found Threshing efficiency was found to be increased rapidly
very slowly decreased with increased of threshing with increase of drum speed with all experimental
sieve size within all the experimental range the mois- range of moisture content, feed rate and threshing
ture content, feed rate and drum speed (Fig. 1b, 1d, 1f). sieve size (Fig. 2b, 2c, 2d). At a fix moisture content of
If we fix the feed rate at 75 kgh-1 threshing, efficiency 5% the threshing efficiency was increased from 88.8 to
was found decreased from 91 to 90% as the sieve size 94.5% as the drum speed increased from 625 to 875
increased from 5 to 9 mm (Fig. 1d). The sieve size was rpm. This may be due to more impact action the chaff
selected for making suitable for all six minor millet were broken down into small parts which were blown
according to their size. Higher size of sieve could not away easily because terminal velocity of chaff de-
give necessary impact and caused grain loss. creased with its reduced mass. Ajav and Adejumo,
Cleaning efficiency: Analysis of variance for response 2005 studied performance evaluation of an okra
surface variable cleaning efficiency is presented in the thresher and got the same result as increase in cylinder
Table 5. The ANOVA data shows that the model is speed increased the cleaning efficiency. Fulani et al.,
significant at 0.1% level of significance whose F value 2013 obtained high cleaning efficiency at higher drum
is 5.8. The linear effect of feed rate on cleaning effi- speed in case of cowpea thresher. Bansal and Lohan
ciency was highly significant (p <0.001). The effect of ( 2009) also found similar result. Cleaning efficiency
drum speed is significant at 5% level of significance. was found slowly decreased as the threshing sieve size
The interaction of drum speed and threshing sieve size increased within all experimental range of drum speed
had significant effect on cleaning efficiency at 0.1% (Fig. 2f) and no effect was found with increased
level of significance. The effect of moisture content threshing sieve size within all the experimental range
and threshing sieve size had no significant effect on the moisture content and feed rate (Fig. 2c, 2e).This
cleaning efficiency. The effect of the interactions ex- may be due to the reason that with increase of drum
cept drum speed and threshing sieve size on cleaning speed threshing efficiency increased and material to be
efficiency were non significant. All the quadratic terms cleaned also increased which reduced the cleaning
of independent parameters had no significant effect on efficiency. If we fix the feed rate at 75 kgh-1cleaning
cleaning efficiency. The regression equation obtained efficiency was found same as the sieve size increased
for cleaning efficiency as the function of four inde- from 5 to 9 mm (Fig. 2c).
pendent variables neglecting the high error generating Optimization of the variables was done by using the
terms was presented in the equation 6. design expert 7.0.0 software by taking threshing and
Ce = +93.50 -0.67 Fr + 1.92 Ds -0.62 Ds x Tss ……..(6) cleaning efficiencies as the two responses. The graphi-
Where, Te = Threshing efficiency, % cal optimization was presented in the Fig 3 a, b, c. By
Mc = Moisture content, % combining the values given in the flagged areas of Fig.
Ds = Drum speed, rev/min 3 a, b and c the optimized values obtained were mois-
Tss = Threshing sieve size, mm ture content 7.79%, feed rate 105.31 kgh-1, drum speed
 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015) 947
of 626.9 rpm, threshing sieve size of 6.0 mm by giving of all minor millets, only the ear heads were fed into
the maximum threshing efficiency of 99.5% and clean- the thresher.
ing efficiency of 88.5%. The values were closer to the
values obtained in the numerical optimization values. Conclusion
On this basis a new thresher was fabricated with the The machine was developed and the independent vari-
optimized values. ables were optimized for maximum threshing and
Testing of the thresher with all the millets: The cleaning efficiency for threshing of all six minor mil-
thresher was operated according to the optimized lets. The optimized values of the independent variables
values of optimized operating parameters. Threshing for maximum threshing efficiency (99.5%) and clean-
of six minor millets was done and threshing capacity, ing efficiency (88.5%) were 7.79% moisture content,
threshing efficiency and broken grain were measured. 105.31 kgh-1 feed rate, 626.9 rpm cylinder speed, 6
The data were analyzed SPSS (v-10) to assess the suit- mm threshing sieve size. The thresher was found suit-
ability of the thresher for the six minor millets and able for threshing of all six minor millets. It was
arranged according to Duncan multiple ranges test adopted successfully by the tribal people for threshing
(DMRT) in the Table 6. The physical parameters of the of all six minor millets.
millets were compared with each other and found sig-
nificant at 5% level of significance. The output with ACKNOWLEDGEMENT
different millet was also found significantly different The authors wish to thank Director, ICAR-Central
at 5% level of significance. Institute of Agricultural Engineering, Bhopal, MP,
The diameter of all the millets was compared with India for providing all necessary requirements for the
each other for the design of sieve. The result was ob- study.
tained that diameter of M1 (2.26 mm), M2 (3.41 mm),
M3 (2.35 mm) and M5 (2.32 mm) are insignificant with REFERENCES
each other and M4 (1.95mm) and M6 (1.63mm) are Abo El-Naga, M.H.M., El-Gendy, H.A. and Mosa, E.H.
insignificant to each other but M1, M2, M3 and M5 are (2015). Evaluation of locally threshing machine per-
significantly different as compared to M4and M6. formance for threshing lentil crop. Misr Journal of Ag-
Sphericity affects the design of outlet and slope of the ricultural Engineering, 22(2): 376-388.
screen for easy gravity flow. The sphericity of M1 Ajav, E.A. and Adejumo, B.A. (2005). Performance evalua-
(0.481) and M4 (0.485) are insignificant to each other tion of an Okra thresher. Agricultural Engineering In-
but significantly different from M2 (0.818), M3 (0.512), ternational: the CIGR Ejournal. Manuscript PM 04006,
M5 (0.544), M6 (0.681). Bulk density is required for the Vol. VII, October.
Anonymous (2001). Grain of truth an analysis. Science and
design of hopper for feeding of crop. The bulk density
Environmental fortnightly magazine, 15 May: 31-41.
of M1 (815.0kg/m3), M4 (825.3kg/m3) and M2 (747.0kg/ Bansal, N.K. and Lohan, S.K. (2009). Design and develop-
m3), M3 (753.7kg/m3) are insignificant to each other ment of an Axial flow thresher for seed crops. Journal
but significantly different from M5 (895.3kg/m3) and of Agricultural Engineering, 46(1):1-8
M6 (802.0kg/m3). Thousand grain weight of all millets Baryeh, E.A. (2002) Physical properties of millet. Journal of
was obtained and compared from which the result was Food Engineering, 51: 39-46.
obtained that M3 (2.39 gm), M4 (2.32), M6 (2.35 gm) Crawford, G.W. and Lee, G. (2003). Agricultural origins in
were insignificant to each other but significantly dif- the Korean Peninsula. Antiquity, 77(295):87-95.
ferent from M1 (2.08gm), M2 (4.20gm) and M5 (3.35 Fulani, A.U., Kuje, J.Y. and Mohammad, M.I. (2013). Effect
of moisture content on performance of a locally fabri-
gm). The threshing capacity was obtained for each
cated cowpea thresher. Journal of Engineering and
millet. Threshing capacity of M1 (88.7 kgh-1), M3(95.3 Applied Sciences, 5(2): 1-15.
kgh-1), M4 (95.0 kgh-1) and M6 (102.0 kgh-1) are insig- Gbabo, A., Gana, I.M. and Amoto, M.S. (2013). Design
nificant to each other but significantly different from fabrication and testing of a millet thresher. Net Journal
M2 (116.7 kgh-1), and M5 (88.7 kgh-1). The threshing of Agricultural Science, 1(4): 100-106.
efficiency of M1 (97.57%), M4 (97.33%) and M5 Goyal, R. K., Vishwakarma, R. K. and Wanjari, O.D. (2008).
(96.33%) are different from threshing efficiency of M2 Optimization of pigeon pea dehulling process. Biosys-
(99.33%), M3(99.40%) and M6 (99.40%). The broken tems Engineering, 99(1): 56–61.
grain percentage of M2 (0.50%), M5 (0.54%) and M6 Kamble, H.G., Srivastava, A.P. and Panwar, J.S. (2003).
Development and evaluation of a pearl millet thresher.
(0.00%)were significantly different from M1 (1.70%),
Journal of Agricultural Engineering, 40(1): 18-25.
M3 (1.20%) and M4 (1.60%). The cleaning efficiency Kushwaha, H.L., Srivastava, A.P. and Singh, H. (2005).
of M1 (97.13%), M4 (96.98%) and M5 (95.89%) are Development and performance evaluation of Okra seed
different from cleaning efficiency of M2(98.96%), M3 extractor. International Commission of Agricultural
(99.12%) and M6 (99.27%). The values of threshing Engineering, 7: 1-13.
and cleaning values of the minor millets after testing Nath, A. and Chattopadhyay, P.K. (2007). Optimization of
were higher than the optimized values. This may be oven toasting for improving crispness and other quality
due to the reason that at the time of testing for optimi- attributes of ready to eat potato-soy snack using re-
zation whole crop was fed into, but during the testing sponse surface methodology. Journal of Food Engi-
neering, 80: 1282 – 1292.
948 K. P. Singh et al. / J. Appl. & Nat. Sci. 7 (2) : 939 - 948 (2015)

Rastogi, N.K., Rajesh, G. and Shamala, T.R. (1998). Optimi- Singh, K.P., Mishra, H.N. and Saha, S. (2010). Optimization
zation of enzymatic degradation of coconut residue. of machine parameters of finger millet thresher-cum-
Journal of Science and Food Agriculture, 76: 129-134. pearler. Agricultural Mechanization in Asia, Africa and
Simonyan, K.J. and Imokheme, P.A. (2008). Development of Latin America 41 (1): 60-67.
an axial flow motorized sorghum thresher. Journal of Singh, K.P., Mishra, H.N. and Saha, S. (2010). Optimization
Agricultural Engineering and Technology, 16(2), De- of Barnyard millet dehulling process using RSM. Agri-
cember: 14-21. cultural Mechanization in Asia, Africa and Latin Amer-
Simonyan, K., Yiljep, Y. and Mudiare, O. (2006). Modelling ica, 41(2), 15-20.
the cleaning process of a stationary sorghum thresher. Singh, K.P., Mishra, H.N., and Saha, S. (2011). Design, de-
Agricultural Engineering International: the CIGR velopment and evaluation of Barnyard millet dehuller.
Ejournal. Manuscript P M 06012, Vol. VIII. August, Journal of Agricultural Engineering, 48 (3): 17-25.
2006. Singh, S.K., Agarwal, U.S. and Saxena, R.P. (2004). Optimiza-
Singh, K.P., Kumar, M., Kumar, A. and Srivastava, A.K. tion of process parameters for milling of green gram
(2008). Effect of wire loop spacing, tip height and (Phaseolusaures). Journal of Food Science and Technol-
threshing drum speed on threshing performances of ogy, 41(2): 124-30.
pedal operated VL paddy thresher. Journal of
Sinha, J.P., Dhaliwal, I.S., Sinha, S.N. and Dixit, A. (2007).
Agricultural Engineering, 45(1),: 12-16.
Studies on machine crop parameters for chickpea seed
Singh, K.P., Kundu, S. and Gupta, H.S. (2002). Vivek
crop threshing. International Commission of Agricul-
thresher for madua and madira. Anonymous, Annual
tural Engineering, 26 (7): 1-9.
report, VPKAS (ICAR), Almora: 88-89.
Singh, K.P., Kundu, S. and Gupta, H.S. (2003). Development Tiwari, B.K., Jagan Mohan, R. and Bhasan, B.S. (2007).
of higher capacity thresher for ragi/kodo. Recent trend Effect of heat processing on milling of black gram and
in millet processing and utilization, CCS, HAU, Hissar, its end product quality. Journal of Food Engineering,
India: 109-116. 78: 356-60.
Singh, K.P., Pardeshi, I.L., Kumar, M., Srinivas, K. and Ushakumari, S.R., Rastogi, N.K. and Malleshi, N.G. (2007).
Srivastva, A.K. (2008). Optimization of machine pa- Optimization of process variables for the preparation of
rameters of a pedal operated paddy thresher using RSM. expanded finger millet using response surface method-
Biosystems Engineering, 100: 591-600. ology. Journal of Food Engineering, 82: 35-42.