Laboratory Evaluation of a Solenoid-Operated Hill Dropping Seed Metering Mechanism for Pre-germinated Paddy Seeds

Abstract

PurposeA precision hill drop seed metering mechanism was developed for sowing of pre-germinated paddy seeds.Methods
The metering mechanism comprised a rectangular metering plate with a hole in the middle, and it was kept at the outlet of the hopper, which was actuated by a push–pull type solenoid to drop the seeds in the form of hill. The triggering of solenoid was controlled by a sensor wheel fitted with an inductive proximity sensor. To reduce choking and bridging of seeds at the hopper outlet, agitation was made by rotating a rod in the hopper. The performance of the developed metering mechanism was evaluated using a grease belt. The effect of main operating parameters—forward speed (0.7, 1, and 1.3 km/h), metering plate hole diameter (9, 10, 11, 12, and 13 mm), and speed of the agitator (20, 40, and 60 rpm)—on the performance of the seed metering mechanism was studied in terms of missing index, multiple index, quality of feed index, mean hill spacing, coefficient of hill distribution uniformity, and coefficient of seed dropping uniformity in hills.ResultsThe multiple index was found to be 0% in all the treatments. No visible damaged seed was found in any treatment. By applying response surface method, the optimum operating condition was found to be 0.84 km/h forward speed, 11.18-mm hole diameter, and 37.37-rpm agitation speed. The missing index, multiple index, quality of feed index, mean hill spacing, coefficient of hill distribution uniformity, and coefficient of seed dropping uniformity in hills at an optimized operating condition were obtained as 6.25%, 0%, 93.75%, 23.5 cm, 90.32%, and 65.22%, respectively. The frequency of occurrence of 3 to 5 seeds per hill was found in 64.8% of the solenoid strokes. The germination percentage of hill drop seeds was 91%.Conclusions
Test results showed that the seed rate variation was much lower than drum seeder, and the number of seeds dropped per hill was closer to the pneumatic rice seeder.

Full text

Vol.:(0123456789)
1 3
Journal of Biosystems Engineering
https://doi.org/10.1007/s42853-021-00124-8
ORIGINAL ARTICLE
Laboratory Evaluation ofaSolenoid‑Operated Hill Dropping Seed
Metering Mechanism forPre‑germinated Paddy Seeds
SujitHensh1 · HiurRaheman1
Received: 8 September 2021 / Revised: 30 October 2021 / Accepted: 7 November 2021
© The Korean Society for Agricultural Machinery 2021
Abstract
Purpose A precision hill drop seed metering mechanism was developed for sowing of pre-germinated paddy seeds.
Methods The metering mechanism comprised a rectangular metering plate with a hole in the middle, and it was kept at the
outlet of the hopper, which was actuated by a push–pull type solenoid to drop the seeds in the form of hill. The triggering
of solenoid was controlled by a sensor wheel fitted with an inductive proximity sensor. To reduce choking and bridging of
seeds at the hopper outlet, agitation was made by rotating a rod in the hopper. The performance of the developed metering
mechanism was evaluated using a grease belt. The effect of main operating parameters—forward speed (0.7, 1, and 1.3km/h),
metering plate hole diameter (9, 10, 11, 12, and 13mm), and speed of the agitator (20, 40, and 60rpm)—on the performance
of the seed metering mechanism was studied in terms of missing index, multiple index, quality of feed index, mean hill spac-
ing, coefficient of hill distribution uniformity, and coefficient of seed dropping uniformity in hills.
Results The multiple index was found to be 0% in all the treatments. No visible damaged seed was found in any treatment.
By applying response surface method, the optimum operating condition was found to be 0.84km/h forward speed, 11.18-mm
hole diameter, and 37.37-rpm agitation speed. The missing index, multiple index, quality of feed index, mean hill spacing,
coefficient of hill distribution uniformity, and coefficient of seed dropping uniformity in hills at an optimized operating
condition were obtained as 6.25%, 0%, 93.75%, 23.5cm, 90.32%, and 65.22%, respectively. The frequency of occurrence
of 3 to 5 seeds per hill was found in 64.8% of the solenoid strokes. The germination percentage of hill drop seeds was 91%.
Conclusions Test results showed that the seed rate variation was much lower than drum seeder, and the number of seeds
dropped per hill was closer to the pneumatic rice seeder.
Keywords Coefficient of hill distribution uniformity· Coefficient of seed dropping uniformity in hill· Precision hill drop
seeder· Quality of feed index· Solenoid
Introduction
Rice establishment methods are broadly classified as dry-
DSR (direct seeded rice), wet-DSR, and transplanting (Kaur
& Singh, 2017). In the wet-DSR method, crop matures early
(8–11days) and also there is saving of cost in terms of labor
and irrigation requirement as compared to the transplanted
rice (Rashid etal., 2009). By using the wet-DSR method,
the steps of transplanting method like establishing the seed-
bed, growing of seedlings, uprooting of seedlings from
seedbed, and transplanting of seedlings in the main field
can be avoided (Bhuiyan etal., 1995). The wet-DSR method
reduces methane gas emission due to no standing water
requirement in the field (Balasubramanian & Hill, 2002).
Moreover, infestation of weeds (Fukai, 2002) and emission
of harmful N2O gas (Ishibashi etal., 2007) are also lesser in
the wet-DSR method as compared to the dry-DSR method.
So, with suitable cultivars, water, and weed management,
it is possible to adapt a direct wet seeding method for rice
cultivation profitably.
In the wet-DSR method, drum seeder is usually used
for line sowing of pre-germinated paddy seeds in wetland
with row to row spacing of 20cm (Rajkumara etal., 2003).
But, there are some major drawbacks for seeding with drum
seeder. The seed rate of drum seeder varies from − 35 to
90% during continuous field operation (Sivakumar, 2001).
* Sujit Hensh
s.hensh1986@gmail.com
1 Department ofAgricultural andFood Engineering, Indian
Institute ofTechnology, Kharagpur, WestBengal721302,
India

Journal of Biosystems Engineering
1 3
Due to this large variation of seed rate, the number of seeds/
hill varies widely. But, the recommended number of seeds/
hill should be within the range of 3 to 5 for optimum yield
(Wang etal., 2014). Moreover, the hill to hill spacings are
not uniform and the seeds scatter widely in the hills. Hence,
thinning or uprooting of growing plants is required to give
sufficient space between plants, as the plant spacing has a
major effect on crop yield. Closer plant spacing affects the
establishment of primary tillers and produces lesser number
of panicles (Huan etal., 1999). Hence, an optimal spacing
is required to be maintained to have more panicles from pri-
mary tillers by minimizing interplant competition. Moreo-
ver, to carry out intercultural operations smoothly, sufficient
plant spacing is required to be maintained. Depending on the
soil fertility and water availability, the recommended plant
spacing normally varies from 15 to 30cm (Anon., 2007). In
the precision hill dropping method, pre-germinated paddy
seeds are sown in wetland by maintaining uniform hill to hill
spacing and number of seeds/hill. Thereby, it reduces seed
rate and increases the yield (Zhang etal., 2018). Several
researches have been conducted on precision hill dropping
of paddy seeds. Khobragade etal., (2012) developed a pneu-
matic seed metering mechanism with a rotating cylindrical
drum. They used Basmati and local (Desi) variety paddy
seeds in this study. The desired seeding with seed place-
ment depth of 0.76 to 0.77cm and hill spacing of 20 to
22cm in puddle soil was achieved at injecting pressure of
1300–1350N/m2 and operating speed range from 0.228 to
0.338m/s. Zhang etal., (2015) designed a precise pneumatic
rice seed metering device with groups of sucking holes. Pei-
zataifeng variety paddy seeds were used in this study. They
found that the probability of dropping 3–4, 2–5, and 5 or
more seeds/hill was 56.13%, 87.6%, and 10.67%, respec-
tively. Furuhata etal., (2015) evaluated the performance of
an air-assisted strip seeder. They used Koshihikari variety
paddy seeds. In this seeder, the seeds were conveyed to
the injection ports by the blower, operated by PTO of the
tractor, and the seeds were deposited in rows with spacing
of 30cm. They found that the planting using air-assisted
strip seeder had more panicles and higher yield than those
using shooting hill seeder. Xing etal., (2017) developed a
pneumatic rice direct-seeder to sow 1 to 3 seeds per hill.
Field experiments were carried out on Peizataifeng variety
paddy seeds. They found that under the optimum condition
of 2kPa negative pressure, the probabilities of getting 1 to
3 seeds per hill were 93.41%, 95.47%, and 97.50% at high
(0.8–0.9m/s), medium (0.5–0.6m/s), and low (0.2–0.3m/s)
forward speed, respectively, and the probability of missing
hill was less than 2%. All the abovementioned researches for
precision hill dropping of paddy seeds are based on pneu-
matic type metering mechanism which requires a blower,
powered by a high-capacity engine or tractor. This leads
to higher operating cost and machine weight as well as
environment pollution. Hence, there is a need to develop
a very simple, cost-effective, and environment-friendly
precision hill drop seed metering mechanism for sowing
of pre-germinated paddy seeds in wetland. Several studies
indicated that precision planter equipped with mechatronics
system provides good seeding uniformity among all seeding
technologies with quality feed index, missing index, multiple
index, and precision index in the range of 90–98%, 0–11%,
0–7%, and 1–22%, respectively, under the travel speed of
1 to 16km/h (Gautam etal., 2019). Singh et al. (2012)
developed a controller-based seed cum fertilizer drill. In the
developed seed drill, the spoked wheel was replaced by a
24-V DC motor operating through pulse width modulation
(PWM). It was tested with soybean seeds. The variation of
seed rate was observed to be 2.16 ± 0.71% as compared to
10–20% observed in the conventional seed cum fertilizer
drills. Sahu (2016) developed a tractor-operated 4-row
inclined plate seed metering planter for cotton seeds with
solenoid-operated shutter valve in all the rows, actuated at
the same time to ensure that seeds be dropped in all the
rows at a time. The switching of solenoids was made by a
sensor wheel (ground wheel with a proximity sensor). Aver-
age seed spacing and standard deviation along the row were
found to be 43.33cm and 2.017cm, respectively. Rajaiah
etal. (2020) conducted a comparative study of mechani-
cal and electronic paddy planters for direct seeding. They
used Pusa-1121 variety dry paddy seeds. The electronic
seed metering unit comprised a DC motor, microcontroller
unit, 16 × 2 char LCD, pulse width modulator (PWM), and
inductive proximity sensor. The observed seed rate, spacing,
and seed placement index values of mechanical and elec-
tronic methods were 22.7kg/ha, 14.4cm, and 74.3% and
19.97kg/ha, 14.8cm, and 86.39%, respectively. Leela and
Saravanakumar (2019) developed an electronically meter-
ized maize planter. The proximity sensor sensed the seed
from the hopper through an optical beam. By using a relay
switch, the solenoid was then triggered linearly to actuate
the seed. An optical driving wheel sensor (encoder) was
used to measure the speed of the ground wheel. Based on
the travel speed information, the speed of the stepper motor
which was driving the metering roll was synchronized to
drop the seeds. The quality feed index was observed to be
94.75%. So, a mechatronics-based metering mechanism may
be one of the options to achieve accurate seed spacing with
higher efficiency in seeding. The above-discussed researches
mainly focused on the mechatronics-based seed metering
mechanism for seeds like soybean, cotton, maize, and dry
paddy. No work has been conducted for the mechatronics-
based hill dropping of pre-germinated paddy seeds. Hence,
a study was undertaken at IIT Kharagpur, India, to develop a
low-cost mechatronics-based precision hill drop seed meter-
ing mechanism for sowing of pre-germinated paddy seeds
and its evaluation for optimum performance.

Journal of Biosystems Engineering
1 3
Materials andMethods
Development oftheSeed Metering Mechanism
The conceptual drawing of the precision hill drop seed
metering mechanism is shown in Fig.1. It consisted of a
frame, hopper, metering plate, push–pull type solenoid,
rotary agitating unit, sensor wheel, electronic circuit, and
12-V battery. The frame was made up of MS (mild steel),
“L” angle of size 20 × 20 × 1.5mm, and adjustable MS flat
of size 35 × 1.5 mm. The hopper was made up of acrylic
transparent sheet for the visibility of seed flow from outside.
The top portion of the hopper was given a square shape. The
bottom portion of the hopper was given an inverted pyra-
mid shape with an inclination angle of 60° for free flow of
paddy seeds; because, the angle of repose of different varie-
ties of paddy varies between 30° and 50° at different levels
of moisture content (Zareiforoush etal., 2009; Patel etal.,
2013). The outlet of the hopper was kept circular of diameter
15mm. To stop the bridging and blocking of seeds at the
hopper outlet, an agitator was provided, which consisted of
a 4-mm-diameter and 150-mm-long MS rod with a trian-
gular-shaped flat piece at the tip as shown in Fig.1, rotated
by a 12-V geared DC motor. The speed of the DC motor
was controlled by an L298N motor driver. The rectangular-
shaped seed metering plate with a hole of specific diameter
at the middle of the plate was placed at the bottom of the
hopper outlet with a clearance of 1mm and supported by an
adjustable MS flat. The metering plate was fixed with the
plunger of a push–pull type 12-V solenoid of stroke length
15mm. The solenoid was mounted on the adjustable MS
flat. The triggering of solenoid was controlled by a sensor
wheel. A specific number of iron rods were placed surround-
ing the center of the wheel. An inductive proximity sensor
(NPN type) was mounted on the wheel’s frame with a clear-
ance of 2mm from the iron rods. During rotation of the
sensor wheel, the iron rods interfered with the proximity
sensor, resulting in an electrical signal, which was sent to
the solenoid through a 12-V relay module. Then, the sole-
noid was actuated and it displaced the seed metering plate
by 15mm for a fraction of second. For this displacement,
the hole of the metering plate and hopper outlet matched
together, which allowed the seeds to flow from the hopper
outlet in the form of hill. Therefore, the metering plate again
returned back to its previous position by the spring force of
the solenoid. The number of hills dropped in one rotation of
the wheel was equal to the number of iron rods placed at the
center of the wheel. The nominal spacing between the hills
was equal to the wheel perimeter divided by the number
of iron rods placed. The electronic circuit of the metering
mechanism consisted of an Arduino Nano microcontroller,
L298N motor driver, 12-V geared DC motor, 5V–10 A relay
module, solenoid, proximity sensor, and 12V–7 Ah bat-
tery. The circuit diagram is shown in Fig.2(a). The L298N
dual H-bridge motor driver was used for speed and direction
control of the DC motor of agitator having a voltage range
between 5 and 35V, with peak current up to 2 A. The relay
module was of SPDT (single pole double throw) type, which
worked as an electromagnetic switch. It consisted of a coil,
one common (COM) terminal, one normally closed (NC)
terminal, and one normally open (NO) terminal. Typically,
the COM and NC terminals had continuity. Whenever the
proximity sensor sensed the iron rod, the coil was getting
Fig. 1 Conceptual drawing of
the precision hill drop seed
metering mechanism. (1) Hop-
per. (2) Seed metering plate.
(3) Push–pull type solenoid. (4)
Agitating rod. (5) Sensor wheel.
(6) Proximity sensor. (7) Iron
rod. (8) Arduino Nano board.
(9) Motor driver. (10) Relay
module. (11) 12-V DC motor.
(12) 12-V battery. (13) MS
frame. (14) Seed delivery tube
1.
Hopper; 2. Seed metering plate; 3. Push-pull type solenoid;4.Agitating rod; 5. Sensor wheel;
6.
Proximity sensor; 7. Iron rod; 8. Arduino Nano board; 9. Motor driver; 10. Relay module;
11. 12 V DC motor; 12. 12 V battery; 13. MS Frame; 14. Seed delivery tube

Journal of Biosystems Engineering
1 3
energized, and the continuity between COM and NO ter-
minals was established and therefore actuated the solenoid.
The working of the entire mechanism was controlled by the
Arduino Nano microcontroller. Arduino IDE software was
used to write the program code and was uploaded in the
Arduino Nano microcontroller. The flow chart of working
of electronic circuit is shown in Fig.2(b).
Laboratory Setup oftheDeveloped Seed Metering
Mechanism
The developed seed metering mechanism was mounted on
a frame above a grease belt of length 4m for conducting
the experiments as shown in Fig.3. The thickness of the
belt was 3mm. The belt was supported by two rollers of
Fig. 2 (a) Electronic circuit
diagram for the seed metering
mechanism. (b) Process flow
chart of the electronic circuit
(a)
(b)

Journal of Biosystems Engineering
1 3
diameter 7cm at the two ends. One roller was driven by a
1.5-hp DC motor through a belt pulley arrangement and the
other roller was free. The speed of the DC motor driving the
grease belt was controlled by a motor speed controller with
AC to DC converter. The driving roller was considered the
sensor wheel during laboratory testing. A small piece of
iron rod of diameter 6mm and length 40mm was mounted
on the extended shaft of the driving roller. A proximity sen-
sor was mounted at a distance of 2mm from the face of
the rod. In one rotation of driving roller, the rod passed the
proximity sensor once and the grease belt surface moved a
distance of (л × (7 + 2 × 0.3)) cm or 24cm (neglecting the
belt slip). That means the nominal hill to hill spacing was
24cm, which was within the recommended range of 15 to
30cm (Anon., 2007).
Research Plan forLaboratory Investigation
The main objective of the laboratory investigations was to
study the effect of operating and machine parameters on
performance of the developed metering mechanism and
also to optimize the parameters to get uniform hill drop-
ping of seeds. The operating parameter in this study was
belt speed or forward speed. The machine parameters were
speed of agitator and metering plate hole diameter. The
measured performance parameters were mean hill spac-
ing, missing index, multiple index, quality of feed index,
coefficient of hill distribution uniformity, and coefficient of
seed dropping uniformity in hills. The definitions of these
indices are explained below.
Missing Index
It is the percentage of spacing between the hills which
are greater than 1.5 times of the theoretical spacing. It
is expressed as follows (ISO 7256/1–1984(E) Standard):
where N = total number of observations and n1 = number of
spacings in the region > 1.5 times of the theoretical spacing.
Multiple Index
It is the percentage of spacing between the hills that are
less than or equal to 0.5 times of the theoretical spac-
ing. It is expressed as follows (ISO 7256/1–1984(E)
Standard):
where N = total number of observations and n2 = number of
spacings in the region less than or equal to 0.5 times of the
theoretical spacing.
Quality ofFeed Index
It is the percentage of spacings that are less than 1.5 times,
but more than 0.5 times of the theoretical spacing. It is
expressed as follows (ISO 7256/1–1984(E) Standard):
(1)
I
miss =
n1
N
×
100
(2)
I
mult =
n2
N
×
100
Fig. 3 Laboratory setup of
the developed seed metering
mechanism. (1) Hopper. (2)
Metering plate. (3) Solenoid.
(4) Agitation unit. (5) Proximity
sensor. (6) Iron rod. (7) Arduino
Nano. (8) L298N motor driver.
(9) Relay module. (10) 12-V
battery. (11) Grease belt. (12)
DC motor. (13) Motor speed
controller. (14) Frame
1.
Hopper, 2. Metering plate, 3.Solenoid, 4.Agitation unit, 5.Proximity sensor, 6. Iron rod, 7.
Arduino Nano, 8. L298N motor
driver, 9. Relay module, 10. 12 V battery, 1
1. Grease belt, 12.
DC Motor, 13. Motor speed controller, 14. Frame

Journal of Biosystems Engineering
1 3
where N = total number of observations, and n3 = number of
spacings between 0.5 times and 1.5 times of the theoretical
spacing.
Coefficient ofHill Distribution Uniformity
The coefficient of hill distribution uniformity was used to
measure the variability in hill to hill spacing. It indicates
closeness of the hill to hill spacings to the theoretical spac-
ing. Higher value of this parameter signifies better uniform-
ity in hill to hill spacings and closer to the ideal spacing.
Mathematically, it is expressed as follows (Bagherpour,
2019):
where Se is the coefficient of hill distribution uniformity (%);
Y is the mean of absolute difference between actual hill spac-
ing and mean hill spacing (cm); and Z is the theoretical hill
spacing (cm).
Coefficient ofSeed Dropping Uniformity inHills
This parameter was used to measure the variability in num-
ber of seeds dropped in hills. It shows the seed dropping
quality in each hill. Higher value of this parameter indi-
cates the number of seeds dropped per hill is closer to the
desired seeds to be dropped per hill. It is expressed as fol-
lows (Maleki etal., 2006):
(3)
I
qf =
n3
N
×
100
(4)
S
e=
(
1−
Y
Z)
×
100
where S is the coefficient of seed dropping uniformity in hills
(%), u is the mean of absolute difference between the actual
numbers of seeds/hill and average number of seeds/hill, and
w is the mean of required number of seeds/hill. The required
number of seeds/hill was 3–5. So, the mean value was 4.
The research plan for the laboratory investigation is pre-
sented in Table1. Full factorial design was used for the
design of experiment. IBM SPSS Statistics 22 software was
used for the statistical analysis and to find the least signifi-
cant difference (LSD) of effective variables. For the opti-
mization of operating and machine parameters, response
surface method was applied by using Minitab 17 software.
In this research, pre-germinated IR36 variety paddy seed
was used. The important physical properties of the seeds
were measured and are summarized in Table2. Dimensions
of 20 numbers of seeds were measured by using a digital
caliper. The geometric diameter (d) of the seeds was calcu-
lated by using the following equation:
(5)
S
h=
(
1−
u
w)
×
100
Table 1 Research plan for
laboratory testing of developed
seed metering mechanism
Sl no Parameters Levels Level value
Common parameters
1 Variety of seed 1 IR36
2 Filling position of hopper 1 Half filling
Independent parameters
1 Belt speed (km/h) 3 0.7, 1, 1.3
2 Hole diameter of metering plate (mm) 5 9, 10, 11, 12, 13
3 Speed of the agitator (rpm) 3 20, 40, 60
Dependent parameters
1 Mean hill spacing
2 Missing index
3 Multiple index
4 Quality of feed index
5 Coefficient of hill distribution uniformity
6 Coefficient of seed dropping uniformity in hills
7 Visible damaged seeds
Table 2 Physical properties of IR 36 variety paddy seeds
Physical properties Minimum Maximum Mean SD
Length, mm 8.64 10.03 9.23 0.37
Width, mm 2.09 2.69 2.36 0.12
Thickness, mm 1.67 2.12 1.97 0.12
Geometric diameter, mm 3.11 3.85 3.5 0.17
Weight of 1000 grains, g 26.85 29.55 27.85 1.4
Bulk density, g/cc 1.26 1.36 1.295 0.065
Angle of repose, ° 38.72 41.06 39.74 1.22

Journal of Biosystems Engineering
1 3
where l is the length, w is the width, and t is the thickness
of the seed.
Experimental Procedure forLaboratory Tests
At first, paddy seeds were pre-germinated. The seeds were
soaked in water for 12h. Thereafter, they were packed in wet
gunny bag and kept in room temperature for 24h for incuba-
tion. Therefore, the seeds were taken out from the gunny bag
and used in experiments. A thin layer of grease was smeared
on the belt to stop bouncing and scattering of falling seeds.
Before performing each test, the hopper was filled half of
its volume with the pre-germinated seeds. The speed of the
grease belt was set to the desired speed by using a motor speed
controller and digital tachometer. The belt speed or forward
speed was calculated by using the following formula:
where V is the belt speed, km/h; D is the diameter of driving
roller, m; and N is the rpm of driving roller of grease belt.
The speed of the motor for rotary agitation was set by
changing the program code in Arduino IDE software and
uploaded in Arduino Nano microcontroller. Five numbers
of seed metering plates with hole diameter 9mm, 10mm,
11mm, 12mm, and 13mm were used in the experiment
(Fig.4). After completion of each test, the number of seeds/
hill and hill to hill spacing of fifteen numbers of consecutive
hills were recorded. The single seeded hills were considered
missing hill; because, the seedling emergence percentage
from pre-germinated seeds varied from 54 to 77% (Hossen
etal., 2018), which means the hills having single seed had 23
to 46% chance of non-emergence of seedling. The number
of damaged seeds was also observed for each test. Each test
was replicated thrice.
(6)
d=(l×w×t)1∕3
(7)
V
=
𝜋×
D
×
N
60
×
3.6
Results andDiscussion
Effect ofForward Speed, Seed Metering Plate Hole
Diameter, andAgitator’s Speed onPerformance
Parameters
The effect of forward speed, metering plate hole diameter,
and agitator’s speed on performance parameters—quality of
feed index (Iqf), missing index (Imiss), multiple index (Imult),
coefficient of hill distribution uniformity (Se), coefficient of
seed dropping uniformity in hills (Sh), and mean hill spac-
ing (Zm)—is shown in Fig.5(a), (b), and (c), respectively.
The corresponding analysis of variance (ANOVA) is shown
in Table3. From Table3, it is found that the forward speed
or belt speed had significant effect on all the performance
parameters at a confidence interval of 99% (p ≤ 0.01) except
Sh, which was found significant at 95% (p ≤ 0.05) level of
confidence. The effect of metering plate hole diameter on
the performance parameters was found significant at 0.99
(p ≤ 0.01) confidence bound. However, no significant effect
of agitator’s speed was observed on the performance param-
eters. Test results revealed that the multiple index was 0%
in all treatments. No visible damaged seed was found in any
treatment. The LSD test (Fig.5(a)) showed that there was
no significant difference in the Iqf, Imiss, Se, Sh, and Zm at
0.7 and 1km/h belt speed at 5% level of significance. But,
significant changes in these performance parameters were
observed at 1.3km/h belt speed. The Iqf and Se decreased,
whereas Imiss, Sh, and Zm increased with the increase of
belt speed from 1 to 1.3km/h. The decrease of Iqf and Se
and the increase of Imiss and Zm with the increase in belt
speed were also reported by Mandal etal. (2018), Rezaei
Asl etal. (2019), and Bagherpour (2019) for vacuumed type
pneumatic seed metering mechanism. The possible reason
could be at higher forward speed, the iron rod crossed the
proximity sensor very fast, which generated poor electrical
signal for actuating of metering plate by the plunger of the
Fig. 4 Seed metering plates of
different diameter holes

Journal of Biosystems Engineering
1 3
solenoid. Thus, the plunger could not complete its full stroke
at higher forward speed. Hence, the probability of falling
of seeds also reduced. In Fig.5(b), it was found that with
the increase in hole diameter of metering plate, the Iqf and
Se increased, and Imiss, Sh, and Zm decreased. Similar find-
ings were also reported by Bagherpour (2019) for vacuum
cylinder precision seeder for chickpea seeds. However, the
LSD test showed no significant difference of performance
parameters at 5% level of significance due to the increase
of hole diameter from 10 to 13mm except in Sh where sig-
nificant variation was observed due to the increase of hole
diameter from 12 to 13mm. The lowest mean hill spacing
Fig. 5 Effect of forward speed
(a), seed metering plate hole
diameter (b), and speed of
agitator (c) on the performance
parameters. Performance
parameters with the same letter
are not significantly different at
p = 0.05 (LSD test)
(a)
(b)
(c)
0
10
20
30
40
50
60
70
80
90
Iqf(%)Imiss(%) Imult(%) Se(%)Sh(%) Zm(cm)
eulav'sretemarapecnamrofreP
0.7 km/h
1 km/h
1.3 km/h
aa
b
aa
b
aa
b
a
a
b
aa
b
0
10
20
30
40
50
60
70
80
90
Iqf(%)Imiss(%) Imult(%) Se(%)Sh(%) Zm(cm)
'sretemarapecnamrofreP
value
9 mm
10 mm
11 mm
12 mm
13 mm
a
bb
b
b
a
b
b
bba
b
b
b
b
a
b
bb
c
a
b
bbb
0
10
20
30
40
50
60
70
80
90
Iqf(%)Imiss(%) Imult(%) Se(%)Sh(%) Zm(cm)
eulav'sretemarapecnamrofreP
20 rpm
40 rpm
60 rpm
aaa
aaa
a
aa
aaa
aaa

Journal of Biosystems Engineering
1 3
was found to be 23.93cm in 0.7–12-40 (belt speed-hole
diameter-speed of agitator) treatment which was closer to
the nominal hill spacing of 24cm. The quality of feed index
varied from 28.6% (in 1.3–9-60 treatment) to 100% (in
0.7–12-40 and 0.7–13-40 treatments). The corresponding
missing indexes were 71.4% and 0%. It showed that “Very
good” seeding quality (Table4) was obtained at 0.7km/h
belt speed, and 12-mm and 13-mm hole diameter. The vari-
ation of coefficient Se was found in between − 30.44% (in
1.3–9-60 treatment) and 98.81% (in 0.7–13-40 treatment).
The negative value of Se indicated that the mean of absolute
difference between actual hill spacing and average hill spac-
ing was more than the nominal hill spacing of 24cm. The
variations of Se and Iqf with belt speed and hole diameter
were similar in nature. However, the coefficient Sh ranged
between − 1.78% (in 1–13-20 treatment) and 91.67% (in
1.3–9-60 treatment). At 1km/h speed and 13-mm hole diam-
eter, the number of seeds/hill was found as high as 23, for
which the mean of absolute difference between the actual
numbers of seeds/hill and average number of seeds/hill was
found higher than the mean of targeted range of 3–5 seeds/
hill. So, the negative value of Sh was obtained. At higher
belt speed of 1.3km/h and smaller size hole diameter of
9mm, the hills formed with the number of seeds closer to
the targeted range of 3–5, but the seeding quality was found
“Insufficient” (Table4) due to the increase of missing index
and hill spacing. Hence, to get the optimized operating con-
dition, a trade-off among the seeding quality and seed drop-
ping uniformity in hills was required.
Optimization ofForward Speed, Hole Diameter,
andSpeed oftheAgitator
The optimization of forward speed, hole diameter, and agi-
tator’s speed was carried out by applying response surface
method, using Minitab 17 software. The goal for Iqf, Se,
and Sh was set maximum. The goal for Imiss and Zm was set
minimum and 24cm, respectively. The result of optimiza-
tion is shown in Fig.6. The optimized values of forward
speed, hole diameter, and speed of the agitator were found
to be 0.84km/h, 11.18mm, and 37.37rpm, respectively.
At the optimized condition, the values of Imiss, Imult, Iqf, Sh,
Se, and Zm were obtained as 11.9%, 0%, 88.09%, 51.55%,
76.94%, and 26.11cm, respectively, with composite desir-
ability of 0.79. After setting the optimized operating condi-
tion (forward speed 0.84km/h, hole diameter 11.18mm,
and agitator’s speed 37rpm), the metering mechanism was
again tested. The values of Iqf, Imiss, Imult, Se, Sh, and Zm at
optimized operating conditions were found to be 93.75%,
6.25%, 0%, 90.32%, 65.22%, and 23.5cm, respectively.
Table 3 ANOVA for effect of belt speed, metering plate hole diameter, and rotary agitation speed on performance parameters
R, speed of agitator; S, belt speed; D, hole diameter; df, degrees of freedom; MS, mean square
* Significant at 5% level
** Significant at 1% level
Source df Iqf Imiss SeShZm
MS F value MS F value MS F value MS F value MS F value
R2 11.92 0.05 11.92 0.05 1018.17 1.34 1.94 0.01 2.91 0.05
S2 1990.72 9.16** 1990.72 9.16** 7959.40 10.47** 1786.50 6.42* 634.54 10.08**
D4 3060.05 14.08** 3060.05 14.08** 7211.93 9.49** 7064.24 25.40** 866.79 13.76**
R × S4 0.39 0.00 0.39 0.00 577.85 0.76 407.00 1.46 0.33 0.01
R × D8 112.61 0.52 112.61 0.52 314.03 0.41 222.32 0.80 96.12 1.53
S × D8 58.63 0.27 58.63 0.27 405.70 0.53 1613.29 5.80* 166.69 2.65
R × S × D16 15.53 0.07 15.53 0.07 10.29 0.01 217.02 0.78 18.97 0.30
Error 45 217.37 217.37 759.93 278.17 62.98
Table 4 Limit values of
performance criteria for
precision seeding (Cay etal.
2018)
Quality of feed index
(Iqf, %)
Multiple index (Imult, %) Miss index (Imiss, %) Classification
> 98.6 < 0.7 < 0.7 Very good
> 90.4 to ≤ 98.6 ≥ 0.7 to < 4.8 ≥ 0.7 to < 4.8 Good
≥ 82.3 to ≤ 90.4 ≥ 4.8 to ≤ 7.7 ≥ 4.8 to ≤ 10 Moderate
< 82.3 > 7.7 > 10 Insufficient

Journal of Biosystems Engineering
1 3
After that, the metering mechanism was run for 10min
continuously with optimized operating condition for each
three filling position of hopper: full, 2/3rd, and 1/3rd to test
the variation of seed rate. The obtained result is shown in
Fig.7. Only 2% variation of seed rate was found between
the full and 2/3rd filling position of hopper. Seed rate vari-
ation of 18% was noticed between 2/3rd and 1/3rd filling of
hopper. Kumar etal. (2017) reported that the seed rate of
drum seeder increased by 131% and 24% due to the change
of filling level of drum from full to half and half to quarter,
respectively. It showed that the variation of seed rate with
filling level of the developed metering mechanism was much
lower than the drum seeder. The metering mechanism was
again tested for 250 strokes of solenoid in optimized oper-
ating condition after filling the hopper up to full level. The
test result is shown in Fig.8. The frequency of occurrence
of 3–5, 3–4, 2–5, and more than 5 seeds/hill was found to
be 64.8%, 53.6%, 78.8%, and 14.4%, respectively. The fre-
quency of occurrence of missing hill and single seeded hill
was found to be 2.4% and 4.4%, respectively. Zhang etal.
(2015) found that the probabilities of dropping 3–4, 2–5,
and 5 or more seeds/hill were 56.13%, 87.6%, and 10.67%,
respectively, in pneumatic rice seeder. It indicated that the
seed dropping quality of the developed metering mechanism
was nearly similar to the pneumatic rice seeder. The sample
images of hill drop seeding by the developed seed metering
mechanism with optimized operating conditions are shown
in Fig.9. To test the germination percentage, hundred num-
bers of randomly selected pre-germinated paddy seeds were
taken from the hill-dropped seeds (Gummert, 2010; Raun
etal., 2002; Vibhuti etal., 2015). Then, the seeds were sown
on a soil-filled tray of size 40cm × 30cm. It was observed
that 91 seeds germinated among the 100 seeds on 10 DAS
(day after sowing). Hence, the germination percentage was
derived 91%.
Conclusions
A solenoid-operated seed metering mechanism was devel-
oped for the precision hill drop seeding of pre-germinated
paddy seeds. The laboratory experiments revealed that the
Fig. 6 Result of optimization in Minitab software. BS, belt speed;
HD, hole diameter of metering plate; SOA, speed of agitator
0
5
10
15
20
25
30
35
40
45
Full 2/3r
d1
/3rd
Seed rate, kg/ha
Filling position of hopper
Fig. 7 Seed rate at different filling positions of hopper with optimized
operating conditions
0
5
10
15
20
25
30
35
40
012345678910
Frequency of occurrence,%
Number of seeds/hill
Fig. 8 Frequency of occurrence vs. number of seeds/hill plot

Journal of Biosystems Engineering
1 3
multiple index was 0% in all the treatments. No visible
damaged seed was found in any treatment. The best seed-
ing quality was obtained at the belt speed, metering plate
hole diameter, and rotary agitation speed of 0.7km/h,
13mm, and 40 rpm, respectively, whereas the highest
seed dropping uniformity in hills was found in 1.3km/h,
9mm, and 60 rpm treatment. The optimized operat-
ing condition was found at a belt speed of 0.84km/h,
plate hole diameter of 11.18mm, and agitator speed of
37.37rpm. At optimized operating condition, the missing
index, multiple index, quality of feed index, coefficient of
seed dropping uniformity in hills, coefficient of hill dis-
tribution uniformity, and mean hill spacing were found
to be 93.75%, 6.25%, 0%, 90.32%, 65.22%, and 23.5cm,
respectively. Variation of seed rate with seed filling level
of the developed metering mechanism was obtained much
lesser than the drum seeder. The frequency of occurrence
of desired seeding (3–5 seeds/hill) was found in 64.8% of
the solenoid strokes. The germination percentage of the
hill drop seeds was obtained as 91%. The performance
of the developed metering mechanism was found closer
to that of the pneumatic rice seeder. Moreover, because
of simplicity in construction, it is very cost-effective for
sowing pre-germinated paddy seeds in wetland as com-
pared to the pneumatic seeder. The developed metering
mechanism is also eco-friendly, as it is operated by battery.
Hence, it is recommended to use in wetland seeder for get-
ting uniform seeding of paddy seeds.
Acknowledgements The facilities provided by the Farm Machinery
& Power Workshop and Laboratory, Department of Agricultural and
Food Engineering, Indian Institute of Technology-Kharagpur, to carry
out this research are sincerely acknowledged.
Declarations
Conflict of Interest The authors declare no competing interests.
References
Anonymous (2007). Rice production manual. http:// www. knowl edgeb
ank. irri. org/ images/ docs/ plant ing- estab lishm ent. pdf. Accessed 20
June 2021
Bagherpour, H. (2019). Modeling and evaluation of a vacuum-cylinder
precision seeder for chickpea seeds. Agricultural Engineering
International: CIGR Journal, 21(4), 75–82.
Balasubramanian, V., & Hill, J. E. (2002). Direct seeding of rice in
Asia: Emerging issues and strategic research needs for the 21st
Fig. 9 Sample images of hill
drop seeding with optimized
operating conditions
Image-1
Image-2
Image-3

Journal of Biosystems Engineering
1 3
century. In S. Pandey, M. Mortimer, L. Wade, T. P. Tuong, K.
Lopez, & B. Hardy (Eds.), Direct seeding: Research strategies
and opportunities (pp. 15–42). International Rice Research
Institute.
Bhuiyan, S. I., Sattar, M. A., & Khan, M. A. K. (1995). Improving
water use efficiency in rice irrigation through wet-seeding. Irriga-
tion Science, 16(1), 1–8.
Cay, A., Kocabiyik, H., & May, S. (2018). Development of an electro-
mechanic control system for seed-metering unit of single seed
corn planters Part I: Design and laboratory simulation. Computers
and Electronics in Agriculture, 144, 71–79.
Furuhata, M., Chosa, T., Shioya, Y., Tsukamoto, T., Seki, M., &
Hosokawa, H. (2015). Developing direct seeding cultivation using
an air-assisted strip seeder. Japan Agricultural Research Quar-
terly, 49(3), 227–233.
Fukai, S. (2002). Rice cultivar requirements for direct seeding in
rainfed lowlands. In: Direct Seeding: Research Strategies and
Opportunities.International Rice Research Institute, Los Ban˜
os, Philippines,(pp. 257–269).
Gautam, P. V., Kushwaha, H. L., Kumar, A., & Kushwaha, D. K.
(2019). Mechatronics application in precision sowing: A review.
International Journal of Current Microbiology and Applied Sci-
ences, 8(04), 1793–1807.
Gummert, M. (2010). Measuring seed germination. http:// www. knowl
edgeb ank. irri. org/ train ing/ fact- sheets/ manag ement- of- other- crop-
probl ems- fact- sheet- categ ory/ measu ring- seed- germi nation- fact-
sheet. Accessed 15 July 2021
Hossen, M. A., Hossain, M. M., Haque, M. E., & Bell, R. W. (2018).
Effect of seed rate on seedling quality for mechanical rice trans-
planting. Bangladesh Rice Journal, 22(1), 9–23.
Huan, T. T. N., Tan, P. S., & Hiraoka, H. (1999). Path-coefficient analy-
sis of direct seeded rice yield and yield components as affected by
seeding rates. Omonrice, 7, 104–111.
Ishibashi, E., Yamamoto, S., Akai, N., & Tsuruta, H. (2007). The influ-
ence of no-tilled direct seeding cultivation on greenhouse gas
emissions from rice paddy fields in Okayama, Western Japan. 4.
Major factors controlling nitrous oxide emission from rice paddy
fields under no-till direct seeding cultivation. Japanese Journal
of Soil Science and Plant Nutrition, 78, 453–463.
ISO. (1984). ISO 7256/1–1984(E): Sowing equipment-test methods e
Part one, single seed drills (precision drills). International Organi-
zation for Standardization.
Kaur, J., & Singh, A. (2017). Direct seeded rice: Prospects, problems/
constraints and researchable issues in India. Current Agriculture
Research Journal, 5(1), 13–32.
Khobragade, H. M., Kamble, A. K., & Dave, A. K. (2012). Performance
evaluation of pneumatic seed metering device for paddy in puddle.
International Journal of Agricultural Engineering, 5(1), 98–102.
Kumar, D., Singh, U. V., Kumar, J., Kumar, M., & Khan, K. (2017).
Laboratory test and calibration of direct paddy drum seeder for
pre- germinated paddy seeds. International Journal of Current
Microbiology and Applied Sciences, 6(9), 1037–1046.
Leela, C., & Saravanakumar, M. (2019). Development of electronically
meterized maize planter. International Journal of Current Micro-
biology and Applied Sciences, 8(04), 2432–2440.
Maleki, M. R., Mouazen, A. M., De Ketelaere, B., & De Baerdemaeker,
J. (2006). A new index for seed distribution uniformity evaluation
of grain drills. Biosystems Engineering, 94(3), 471–475.
Mandal, S., Kumar, G. V. P., Tanna, H., & Kumar, A. (2018). Design
and evaluation of a pneumatic metering mechanism for power
tiller operated precision planter. Current Science, 115(6),
1106–1114.
Patel, N., Jagan, S. K., Jha, S. K., Sinha, J. P., & Kumar, A. (2013).
Physical properties of Basmati varieties of paddy. Journal of Agri-
cultural Engineering, 50(4), 39–47.
Rajaiah, P., Mani, I., Parray, R. A., Lande, S. D., Kumar, A., &
Vergese, C. (2020). Design and development of precision planter
for paddy direct seeding. Journal of Agricultural Engineering,
57(4), 302–314.
Rajkumara, S., Hanamaratti, N. G., & Prashanthi, S. K. (2003). Direct
wet seeding in rice - A review. Agricultural Reviews, 24(1), 57–63.
Rashid, M. H., Alam, M. M., Khan, M. A., & Ladha, J. K. (2009).
Productivity and resource use of direct-(drum)-seeded and trans-
planted rice in puddled soils in rice–rice and rice–wheat ecosys-
tems. Field Crops Research, 113, 274–281.
Raun, S., Xue, Q., & Thlkowska, K. (2002). Effect of seed priming on
germination and health of rice (Oryza sativa L) seeds. Seed Sci-
ence and Technology, 30, 451–458.
RezaeiAsl, A., Roudbari, M., & Esmailzadeh, E. (2019). Fabrication
and evaluation of vacuumed metering drum performance for row
planting of soybean with grease belt. Agricultural Engineering
International: CIGR Journal, 21(4), 96–106.
Sahu, G. (2016). Development and evaluation of simultaneous seed
dropping mechanism for planter. M.Tech. thesis. IIT Kharagpur,
India
Sivakumar, S.S. (2001). Investigation on the performance evaluation
of drum seeder as influenced by machine, ergonomic and crop
parameters. Ph.D. thesis. Tamil Nadu Agricultural University,
India
Singh, K., Agrawal, K. N., Dubey, A. K., and Chandra, M. P. (2012).
Development of the controller based seed cum fertilizer drill.
12th International Conference on Intelligent Systems Design and
Applications (ISDA), 369–374. https:// doi. org/ 10. 1109/ ISDA.
2012. 64165 66
Vibhuti, Shahi, C., Bargali, K., & Bargali, S. S. (2015). Seed ger-
mination and seedling growth parameters of rice (Oryza sativa)
varieties as affected by salt and water stress. Indian Journal of
Agricultural Sciences, 85(1), 102–108.
Wang, D., Chen, S., Wang, Z., Ji, C., Xu, C., Zhang, X., & Chauhan,
B. S. (2014). Optimizing hill seeding density for high-yielding
hybrid rice in a single rice cropping system in South China. PLOS
ONE, 9(10), 1–14. https:// doi. org/ 10. 1371/ journ al. pone. 01094 17
Xing, H., Wang, Z. M., Luo, X. W., Cao, X. M., Liu, C. B., & Zang,
Y. (2017). General structure design and field experiment of pneu-
matic rice direct-seeder. International Journal of Agricultural and
Biological Engineering, 10(6), 31–42.
Zareiforoush, H., Komarizadeh, M. H., & Alizadeh, M. R. (2009).
Effect of moisture content on some physical properties of paddy
grains. Research Journal of Applied Sciences, Engineering and
Technology, 1(3), 132–139.
Zhang, M., Wang, Z., Luo, X., Zang, Y., Yang, W., Xing, H., Wang,
B., & Dai, Y. (2018). Review of precision rice hill-drop drilling
technology and machine for paddy. International Journal of Agri-
cultural & Biological Engineering, 11(3), 1–11.
Zhang, G. Z., Zhang, S. S., Yang, W. P., Lu, K., Lei, Z. Q., & Yang,
M. (2015). Design and experiment of double cavity side-filled
precision hole seed metering device for rice. Transactions of the
CSAE, 32(8), 9–17.