DEVELOPMENT OF A TRACTOR MOUNTED AUTOMATIC VEGETABLE

TRANSPLANTER

Thesis submitted in part fulfillment of the requirements for the award of degree of
DOCTOR OF PHILOSOPHY IN FARM POWER AND MACHINERY
to Tamil Nadu Agricultural University, Coimbatore

by
S.SIVAKUMAR, M.E. (Agri.,)
ID.NO. 10-643-002

DEPARTMENT OF FARM MACHINERY

AGRICULTURAL ENGINEERING COLLEGE AND RESEARCH INSTITUTE
TAMIL NADU AGRICULTURAL UNIVERSITY
COIMBATORE – 641 003

2014
DEVELOPMENT OF A TRACTOR MOUNTED AUTOMATIC VEGETABLE
TRANSPLANTER

by
S.SIVAKUMAR, M.E (Agri.,)
ID.NO. 10-643-002

DEPARTMENT OF FARM MACHINERY

AGRICULTURAL ENGINEERING COLLEGE AND RESEARCH INSTITUTE
TAMIL NADU AGRICULTURAL UNIVERSITY
COIMBATORE – 641 003

2014
 CERTIFICATE

This is to certify that the thesis entitled “DEVELOPMENT OF A TRACTOR

MOUNTED AUTOMATIC VEGETABLE TRANSPLANTER” submitted in part
fulfillment of the requirements for the award of degree of DOCTOR OF PHILOSOPHY in
FARM POWER AND MACHINERY to the Tamil Nadu Agricultural University,
Coimbatore, is a record of bonafide research work carried out by Er. S.SIVAKUMAR,
ID.NO. 10-643-002, under my supervision and guidance and that no part of the thesis has
been submitted for the award of any other degree, diploma, fellowship or similar title or
prizes and that the work has been published in part or full in any scientific or popular
journal or magazine.

Place : Coimbatore (Dr. C.DIVAKER DURAIRAJ)

Date : Chairman

Approved by :

Chairman : (Dr. C.DIVAKER DURAIRAJ)

Co opt members : (Dr. D.ANANTHA KRISHANAN)

(Dr. B.SHRIDAR)

Member : (Dr. L.PUGALENDHI)

Date (EXTERNAL EXAMINER)

(G.ARAVIND REDDY)
Acknowledgement
 ACKNOWLEDGEMENT

I deem it as a great pleasure and privilege to place on record my profound sense of

gratitude, indebtedness and heartful thanks to chairman of the advisory committee
Dr.C.Divaker Durairaj, Dean (Engg), i/c, Agricultural Engineering College and Research
Institute, Coimbatore for rendering incessant guidance, cooperation, efficacious advice
and giving valuable suggestions throughout the course of this investigation and in the
preparation of the thesis.

I owe a great debt to the honorable members of the advisory committee

Dr.V.J.F.Kumar, Professor , Department of Farm Machinery, AEC & RI, Kumulur,
Dr.D.ManoharJesudas, Professor, Department of Farm Machinery, AEC & RI,
Coimbatore, Dr.L.Pugalendhi, Professor (Vegetable crops) , TRS, Yathapur for their
laudable counseling, illuminating suggestions and constructive criticisms.

I acknowledge with gratitude and sincere thanks to Dr.D.Asokan, Professor and

Head, Department of Farm Machinery, for providing me with all the necessary facilities
for the smooth conduct of research work.

I wish to register my indebtedness to Dr.D.Ananthakrishnan, Dr.K.Kathirvel,

Dr. V.M.D.Duraisamy, Professors, Dr.V.Kumar, Professor (SWCE), Dr.B.Shridar,
Professor and Head, AMRC, Dr.A.Surendrakumar, Professor, AMRC and
Dr.M.Saravanakumar, Dr.V.Palaniselvam Assistant Professor, Department of Farm
Machinery for their unreserved help, incessant motivation and encouragement during the
tenure of my study.

I place my bountiful thanks to my Ph.D classmates Er. Balachand CH,

Er.A.P.Mohankumar and Er.K.Baladhandabhani, Er. M.M.C.Rajeevgandhi for
their affection and encouragement during the study.

I am short in words to acknowledge the selfless services of Mr.K.Nagarajan,

Mr.C.Saravanakumar, Mr.A.Dhandapani, Mr.P.Natchimuthu, Mr.S.Poovendran,
Mr.P.Thirumoorthy, Mr.S.Krishnamoorthy, Mr.S.MohamedYousuf, Mr.N.Mahendra
kumar, Mr.R.Vadivelu, Mr.R.Murugan, Mr.C.N.Muthuselvam, Mrs. S.Thilagavathy
Mr.G.Balakrishanan, S.Yasodha Krishnan, B.Yuvaraj, Sasikumar, Mr. S.Raja,
Mr.A.R.Christy allen, Sivasubramaniam, artisans and Mr.N.Kamaraj, Superidentent for
their participation in the study and sincere helps by all other artisans of the workshop is also
gratefully acknowledged.

Words are inadequate in offering my thanks to Er. Sreerangurayar, Er.Prakash,

Er.K.Priya, J.R.F, P.Mallika, G.Bagiyalakshimi, P.Selvam for the timely support and
service in the study.

Sincere thanks to Mr. G.Sathiyamoorthi and Mr. N. Subramaniam for their help
rendered in Auto- CAD and Solid works software skills.

I express my sincere gratitude to my department friends Er. M.Monoharan,

Er.Edwin Benjamin, Er. P. Tamilselvi, Er.G. Vasuki, Er.D. Ganapathy,
Er.S.Syed Imran, Er.A.N.Rajesh, Er.Navaneethapandian, Er.Vinothkumar,
Er. K.Kartikeyan and I PG, II PG students, Department of Farm Machinery for their
timely help rendered to me during my study period.

I am thankful to Dr. M.Manikandan, Dr.G.Vijayakumar, Dr. V.Dharmaraj,

Dr. R.Mythili, and Dr.P.Dhananchezhiyan for their support and help rendered during
my post graduation studies.

I wish to express my thanks to M/s. Amman nursery, Thondamuthur,

M/s.kumaran Xerox for timely support and service in the study.

I am grateful to Dr.B. Suthakar, Dr.P.Kamaraj, Dr. S.Sriramajayam, and

Dr.D.Ramesh Assistant Professors for their constant encouragement throught my
educational carrier without they the work will be an intricate one.

I grateful acknowledge Council of Scientific and Industrial Research, New Delhi

for the fellowship given to pursue my doctoral studies.

I am greatly indebted to the Tamil Nadu Agricultural University for providing an

opportunity to undergo this doctoral programme.

(S.SIVAKUMAR)
Abstract
 ABSTRACT

DEVELOPMENT OF A TRACTOR MOUNTED AUTOMATIC VEGETABLE

TRANSPLANTER

By

S.SIVAKUMAR

Degree : Doctor of Philosophy (Farm Power and Machinery)

Chairman : Dr. C. DIVAKER DURAIRAJ

Dean (Engineering) i/c
Agricultural Engineering College and Research Institute
Tamil Nadu Agricultural University
Coimbatore - 641 003

2014

Vegetable transplanting is presently carried out only manually in India. In the

process of manual transplanting, roots are severely damaged and the plants take longer
time to establish their roots. Hence vegetable growers have moved on to manual
transplanting of protray seedling.

Mechanization of vegetable transplanting is the need of the hour. The currently

developed transplanters in India are semi automatic. They are not suitable for continuous
operation over a long period of time. This work was carried out to systematically design
and develop an automatic vegetable transplanter suitable for protray seedlings. In
developing an automatic vegetable transplanter, the foremost factor is the seedling
removal which needs to be automated.

Three methods of plant removal such as gravity falling, clawing out from the top
of the protray cell and popping out the seedling were developed and tested in the
laboratory. The gravity falling and clawing out method of plant removal were discarded
due to of inherent difficulties. A slider crank mechanism was developed and motorized
to pop out the seedling as pushed from the bottom of the protray.

The ejection performance was assessed as influenced by age of seedling, growth

media, plunger diameter and speed of travel through logit analysis. For a 6 mm diameter
plunger, the probability of successful ejection was more than 90 per cent for all the media
with 20 days old seedling.

A 75 per cent shaded net house was built for raising plug transplants with different
growth media. A rack system was also developed for growing the plug transplants in
constructed shaded house, to prevent root propagation through the bottom of tray.

The plant growth as influenced by age and growth media was assessed in terms of
plant height, stem diameter, number of leaves, plug weight and root weight. The
seedlings grown in coir pith + vermicompost (4:1) provides the best results. The
maximum height of 230 mm with a stem thickness of 3.6 mm and the number of leaves
were found to be high in 40 days old seedling.

The ejected seedling, when falling down to the ground directly or on a trap door
during transplanting, may get damaged in the root ball. A drop test was conducted to find
the root weight loss on impact of plug to floor. The loss was found to be lesser than
5 per cent with cushion of 40 and 20 mm when dropped from 250, 500 and 750 mm
heights.

It is imperative that the horizontally ejected seedling should straighten when it

reaches the ground level. An experiment was conducted to determine the straightening
height of the free falling plant. It is observed that the probability of straightening is
70 to 80 per cent at 500 and 750 mm height for the different media tested.

A prototype vegetable transplanter was developed and the transplanting variables

on planting performance were optimized. The results showed that the moisture content of soil
is to be at 10 per cent (w.b), the compaction wheel angle 15 , depth of planting 75 mm and
speed of travel 0.5 km h-1 for optimal performance in the field.

An automatic transplanter mechanism was devised next which had the following
constituent parts.
 * A horizontal rail and slider assembly carrying the vertically loaded seedling
protray, which slides along the travel direction of the machine

* A vertical column of seven ejector mechanisms that are positioned so as to eject

plugs from one single vertical column of the protray at a time

* A motorized walking beam mechanism to move the tray slider in precise steps
corresponding to the inter distance between cells on the protray.

* Optical detector for precise positioning of the protray cells relative to the ejector
plungers.

* Main microcontroller (PIC16F128) board controlling the logic of tray movement

and plug ejections in proper sequence.

The main microcontroller board provides the logic for the automatic sequence of
operation of the machine. It starts to sequentially operate seven ejectors from the bottom
to the top thus ejecting one column of seedlings from the protray. It then moves the tray
to the next column for ejection. After completing 14th column, a new tray is fed and
again the ejection starts from last column to first column of the protray and so on.

After the incorporation of the automatic transplanting mechanism on the prototype

transplanter, a field test was conducted to assess the performance of the machine.

The transplanter placed the seedling accurately in the furrow at the expected
spacing of 450 mm with a variation of ± 25 mm and a success ratio of 80 per cent.

The field capacity and field efficiency of the machine was found to be
-1
0.015 ha h and 50 per cent respectively. The cost of operation of the automatic
vegetable tranplanter was Rs. 2072 ha-1, which was less than conventional method. The
transplanter resulted in 97 per cent saving in labour and 90 per cent in labour cost. The
cost of seedling was Rs.23, 187 ha-1. The cost of the automatic vegetable transplanter was
approximately Rs. 25,000.

As a whole, this work was able to develop a working model of an automatic

vegetable transplanting machine which can eject the seedling from a protray and plant
them.
 CONTENTS

CHAPTER PAGE
TITLE
NO NO.

LIST OF TABLES i

LIST OF FIGURES ii

LIST OF PLATES vi

LIST OF SYMBOLS AND ABBREVIATIONS viii

I INTRODUCTION 1

II REVIEW OF LITERATURE 7

III MATERIALS AND METHODS 51

IV RESULTS AND DISCUSSION 119

V SUMMARY AND CONCLUSIONS 180

REFERENCES
 LIST OF TABLES

Table Page
Title
No. No.

2.1 Labour requirement for transplanting of selected vegetable crops 10

2.2 Field planting rates with semi automatic transplanters 11

Characteristic of tomato plants grown in blackmore square cell and

2.3 33
round cell

2.4 Field performance of two row semi automatic transplanter 48

3.1 List of selected variables for seedling ejection 78

3.2 List of selected variables for agronomical parameters 80

3.3 List of selected variables for root ball weight loss analysis 86

3.4 List of selected variables for straightening height 86

3.5 List of selected variables for field level optimization by manual 94

dropping

4.1 Analysis of variance on seedling height 130

4.2 Analysis of variance on stem thickness 131

4.3 Analysis of variance on no of leaves 133

4.4 Analysis of variance on plug weight 134

4.5 Analysis of variance on stem weight 143

4.6 Analysis of variance on root weight 147

4.7 Analysis of variance on pulling force 153

i
 LIST OF FIGURES

Figure
Title Page No.
No.

3.1 Open bottom protray segment (1/7 th of a full tray) 55

3.2 Transfer plate 55

3.3 Test rig for seedling ejection by gravity 57

3.4 Circuit diagram for imparting movement to the protray segment 58

3.5 Double crank with belt drive 62

3.6 Paper cone 64

3.7 Set up for clawing out the plug from top of protray 66

3.7.a Finalized coupler curve of the mechanism 66

3.8 Tray holder 67

3.9 Slider crank mechanism 69

3.10 Motorized slider crank mechanism 72

3.11 Regenerative stopping of ejector motor - circuit 75

3.12 Rack for growing seedling 82

3.13 Pulling force measurement assembly 82

3.14 Main frame 89

3.15 Hopper 91

3.16 Furrow opener 91

3.17 Compaction wheel 92

3.18 Protray slider assembly 103

3.19 Slider 103

3.20 Assembly of the automatic ejection mechanism 104

ii
Figure
Title Page No.
No.

3.21 Plant ejector assembly 105

3.22 Ejector motor drive relay board 109

3.23 Main microcontroller board 110

3.24 Ground wheel assembly 111

3.25 Flow diagram 113-115

Schematic diagram for working of the prototype automatic

3.26 116
vegetable transplanter

3.27 Prototype automatic vegetable transplanter 117

4.1 Effect of gravity falling with open bottom type protray 120

4.2 Effect of medium on plant ejection 123

4.3 Effect of age on plant ejection 123

4.4 Effect of different media on seedling height 132

4.5 Effect of different media on stem thickness 132

4.6 Effect of different media on number of leaves 133

4.7 Diagnostics on model of plug weight (first model tried) 135

Determination of Box - Cox transformation parameter (plug

4.8 136
weight)

4.9 Diagnostics on model of plug weight (model appropriate) 137

Tukey mean comparison on the root weight of nursery plants as

4.10 138
influenced by growth medium

4.11 Effect of plug weight on different medium and age 139

4.12 Diagnostics on model of stem weight (first model tried) 143

Determination of Box - Cox transformation parameter (stem

4.13 144
weight)

iii
Figure
Title Page No.
No.

4.14 Diagnostics on model of stem weight (model appropriate) 145

Tukey mean comparison on the stem weight of seedling as

4.15 146
influenced by growth medium and age

4.16 Effect of stem weight on different medium and age 146

4.17 Diagnostics on model of root weight (first model tried) 148

Determination of Box - Cox transformation parameter (root

4.18 149
weight)

Tukey mean comparison on the root weight of nursery plants as

4.19 149
influenced by growth medium

4.20 Diagnostics on model of root weight (model appropriate) 150

4.21 Effect of plug weight on different medium and age 151

4.22 Diagnostics on model of pulling force (model appropriate) 152

Tukey mean comparison on the pulling force of nursery plants

4.23 153
as influenced by growth medium and age

4.24 Effect of pulling force on different medium and age 155

Diagnostics on model of plug weight loss due to fall (model

4.25 156
tried)

Diagnostics on model of plug weight loss due to fall model

4.26 157
appropriate after square root transformation

Influence of cushion thickness, media and height of fall on

4.27 159
weight loss due to fall

a Effect of cushion and height of falling 159

b Effect of cushion and plug medium 159

Influence of media, days of hardness and cushioning on plug

4.28 161
weight loss due to fall

a Effect of cushion and hardening 161

iv
Figure
Title Page No.
No.

b Effect of height of falling and plug medium 161

Effect of media, days of hardening and height of fall as weight

4.29 162
loss due to falls

a Effect of hardening and height of falling 162

b Effect of hardening and plug medium 162

4.30 Tukey mean comparison on the plug weight loss due to fall 163

4.31 Effect on height of fall on success of straightening 171

Probability of successful uprightness due to the effect of

4.32 compaction wheel angle and depth of planting with forward 173
speed of 0.5 km h -1

a Depth 50 mm 173

b Depth 75 mm 173

c Depth 100 mm 174

Probability of successful uprightness due to the effect of

4.33 compaction wheel angle and depth of planting with forward 175
speed of 1 km h -1

12a Depth 50 mm 175

b Depth 75 mm 175

c Depth 100 mm 176

Probability of successful uprightness due to the effect of

4.34 compaction wheel angle and depth of planting with forward 177
speed of 1.5 km h -1

a Depth 50 mm 177

b Depth 75 mm 177

c Depth 100mm 178

v
 LIST OF PLATES

Plate
Title Page No.
No.

3.1 Open bottom type pro - tray (7 segments make one full tray) 56

3.1.a Open bottom pro - tray segment 56

3.1.b Transfer plate 56

3.2 Experimental test rig for seedling ejection by gravity 59

3.3 Finger wheel moving mechanism 59

3.4 Walking beam mechanism with belt drive 63

3.5 Growing of seedling in open bottom pro - tray segments 63

3.6 Four bar mechanism 70

3.7 Slider crank mechanism 70

3.8 Pro - tray holder 73

3.9 Motorized slider crank mechanism 73

3.10 Shaded net house exterior view 83

3.10.a Shaded net house-inside view 83

3.11 Rack with pro tray seedling 83

3.12 Pulling force measurement assembly 84

3.13 Experimental set - up for drop test 84

3.14 The tray seedling with binding materials 87

3.15 Experimental set up for plant straightening height 87

3.16 Furrow opener 93

3.17 Compaction wheels 93

vi
Plate
Title Page No.
No.

3.18 Tray slider assembly 106

3.19 Slider 106

3.20 Plant ejector assembly 107

3.21 Main microcontroller board 107

3.22 Opto sensor 108

3.23 Relay board assembly 108

3.24 Ground wheel 112

3.25 Hopper 112

3.26 Prototype automatic vegetable transplanter 118

3.27 Transplanted field 118

vii
LIST OF SYMBOLS AND ABBREVIATIONS

cc - Cubic centimeter

ccc - Cholor mequat cloride

cm - Centimeter

DC - Direct current

 - Degree

df - Degree of freedom

et al - And others

etc - Etcetera

Fig. - Figure

FYM - Farm yard manure

g - Grams

h - Hour

ha - Hectare

ha h-1 - hectare per hour

kg - Kilogram

km h-1 - kilometer per hour

lb - Pound

m - Metre

viii
min - Minute

mm - Millimetre

m s-1 - Metre per second

N - Newton

ppm - parts per million

PTO - Power take off

PVC - Pilyvinyl chloride

Pr - Probability

rpm - Revolution per minute

Rad - Radian

Rs - Rupees

s - Second

ss - Sum of square

V - Volt

v/v - Vilume basis

viz - Namely

w.b - Wet basis

± - Plus or minus

@ - At the rate of

ix
Introduction
 CHAPTER I

INTRODUCTION

India has emerged as a leading horticultural country of the world with a total annual
production of 240 million tonnes of horticultural crops during 2010 - 11. Developing
countries contribute 72 per cent of the total vegetable production in the world. India is the
second largest producer of vegetables after China. Production of vegetables in India stands at
14.0 per cent of the world production in 2010 - 11. India produces nearly 60 leafy, fruity and
tuber varieties of vegetables. The area under vegetable is around 8.50 million ha and
production 146.55 million tones (Anonymous, 2011). Most of the farm operations except
seedbed preparation are carried out manually. Therefore, vegetable crops are usually
grown commercially, where the labour is available at low cost. The transplanting
operation is one of the most labour intensive operations in vegetable production. It is
largely done by hand in India and incurs large investments in labour, time and
cost (Kumar and Raheman, 2008). Each field worker is only able to set 400 plants h-1
(Suggs et al., 1989), so a considerable labour force is demanded. Moreover, there are
several activities involved in the vegetable transplanting operation. These include
preparing the field for placing the seedling as well transporting the seedling from the
nursery to the field. So vegetable farming needs to discover novel technology for growing
uniform size and healthy seedling. Hence the vegetable growers have moved on to the
protray seedling. The protray seedling has good reestablishment capability after
transplanting them manually or mechanically.

1.1 Conventional method of vegetable nursery growing

Vegetables like tomato (Solanum lycopersicum), egg plant (Solanum melongena)

and pepper (Capsicum spp.) are first sown in nursery bed and later transplanted manually
either on ridges or on well prepared flat beds. Raising the seedlings takes a relatively long
time. A large amount of medium is used and since the components of the medium are
neither uniform nor sterile, the result may be diseased on seedlings of different quality
and size. A relatively large area of land is required for growing and management of
seedling and the costs are high. The result is poor transplant quality and high production
cost. The production by individual growers of their own transplants has some limitations.
1
The low level of technology used, the small scale of production and the lack of
standardized materials or a standardized production environment are not favourable for
the production of high quality transplants. Eventhough labour input is high, the
transplants are poor in quality.

1.2 Conventional method of vegetable transplanting

Vegetable transplanting at present is carried out manually all over the country.
While transplanting is done manually, roots are severely damaged in the process of
transplanting. So the plants take longer to establish their roots after transplanting.
Transplanting of vegetable seedling manually is very tiresome and labour consuming as the
operation is done in a bending posture. High labour requirement and shortage of labour
during peak transplanting season causes delay in transplanting and affects timely operation.

A survey was carried out to assess mechanization gaps in raising seedling, planting
and transplanting of vegetable crops. The results of the survey indicated that there was a
mechanization gap in area of transplanting of vegetable crops (Chaudhuri et al., 1999).

1.3 Protray growing of nursery

Plug or cell transplants are seedling or small seed propagated plants which are
raised in individual small cells, called plugs. Plug trays or protrays are available in
plastic and Styrofoam which have an array of cells. The plugs are filled with a cohesive
medium, and are eventually transplanted into other growing systems. Seeds are sown
manually or by automated seeders into plug trays. Only one plant is raised per cell. The
protray grown transplants has been widened to provide a stable supply of high quality
seedlings and improved transplanting efficiency (Byoung Ryong Jeong, 1992).

Compared to traditional nursery beds, seedlings that are raised using protray
germinate early and are vigorous. There is good root development in the seedling. Low
mortality rates of the seed is observed that are sown in protray. Protrays are presently
used for growing most of the vegetables and all kinds of flowers. Trays are available with
optional top surface ventilation holes, precisely centred punched drain holes, and cell
wall root ribs to promote downward root grown. The protrays are available in the market
in different sizes and shapes. The cell is round, square and truncated inverted pyramid

2
respectively. The trays contain 128, 200 or 288 cells in which the arrangement of cells is
8 x 16, 10 x 20 and 12 x 24. The most popular one being used in India has 14 x 7 cells
with a cell volume of 28.19 mm3.

The use of transplants may actually reduce the cost of establishing vegetable
planting because less quantity of seed is used and the need for thinning and early weeding
is eliminated. The transplants also allow more efficient use of fertilizer and irrigation
water during early growth stages. The successful production of transplants requires
sterilized growth media, temperature, light control, effective pest control and disease
management; and appropriate sanitation practices. The protray grown seedlings are
individual in each cell and this prevents root entanglement with neighbouring plants.
Very aged seedlings may have roots propagated from the cell to next through the top.

The greatest damage occurs when bare root transplants are pulled from the soil and
when roots are intentionally pruned off. Protray transplant production systems minimize
the disturbance of root to soil contact, but some root damage does occur when the plant
are pulled from the trays. The plug type nursery has gained popularity in India with
almost all the vegetable farmers. Lot of entrepreneurs have put up net houses, exclusively
for supply of protray grown seedlings.

The plug system, in particular, has the operation systematized, integrated and
mechanized, producing highly uniform transplants. Vegetable crop growers simply buy
the transplants of their choice from the plug growers and devote their efforts to growing
their crops rather than nursery. Plug plants are high in quality as well as uniform. This
makes mechanized transplanting possible thus reducing production cost. The supply of
professionally produced high quality transplants is plentiful, stable, and timed so as to
meet grower’s need. Handling and transportation from the net house to field is easy.

The tray grown plants cost more but their survivability is better and transplanting
shock is much less. The tray grown seedling have stirred up interest in developing fully
automatic transplanters, since the seedling are in an orderly array that might be handled
mechanically (Shaw, 1999).

Moreover, vegetable growers prefer protray grown seedlings over conventional

ones even for manual transplantation itself. The reasons are that the process of producing
3
virile nursery in their farms is avoided, eliminating the underlying risks and labour
involvement. The tray seedlings are grown under controlled environments and have
appreciable vitality and plant stand after transplantation. It is reasoned that all the vital
inputs to the growing seedling are rendered precisely in a regulated way so as to allow
them to acquire these attributes. The handling of seedling during transplantation is
easened considerably.

1.4 Vegetable transplanters

The currently available transplanters are divided into two categories viz., fully
automatic transplanter and semi automatic transplanter. The semi automatic transplanter
has a limited operating speed due to the need for manual feeding of one seedling cell after
another. It is not suitable for continuous operation over a long period of time. The fully
automatic transplanter will allow for relatively high speeds of operation and labour
saving because seedling will be fed automatically by the machine itself. However fully
automatic transplanters that have been developed so far internationally are limited and
very scarce in India.

These machines are further divided into walking type and riding type. In addition,
the type of transplanter may vary depending on the seedling to which it can adapt. The
seedling used for transplanting by a machine can be broadly divided into seedling with
soil and seedling without soil (bare root seedling). Seedling with soil can be further
divided into cell mold seedling, paper pot seedlings, kneaded nursery seedling and soil
block seedlings.

Riding type two row or three row semi automatic transplanter are tractor mounted
or tractor pulled machines. The riding type transplanter has seats for labour, who feed
seedlings to the planting unit, while the automatic transplanter has a seedling pickup
device to mechanically transfer seedling from tray to the planting unit. In the fully
automatic transplanter, the tray is kept at a place with its orientation suitable for the
pickup device to remove seedling from trays. It also has a mechanism to move trays
forward as seedlings are removed from the tray.

In India, some semi automatic transplanters have been developed, where the
feeding of seedling is done manually and metering mechanically. The field capacity of
4
these units is however restricted by the feeding capacity of labour. The transplanter needs
to be operated at a very slow speed so that the operator will have time to pick the seedling
and position it for transplanting. This makes the coverage very meagre for practical
applications. Moreover, at the prevailing labour wages in rural areas at Rs.150 - 200 day -1
which is likely to go up in future, the cost of labour will become prohibitively higher.
Thus there exists a dire need to introduce automatic vegetable transplanters for
mechanical transplanting of vegetable.

1.5 Need of automatic vegetable transplanter

Mechanized transplanting of vegetable is not widely practised in India and most of

the developing countries. The plant removal and plant setting is important factors to be
considered while designing automatic transplanters. The main problem for researchers in
developing a fully automatic transplanter, is the technology of how to remove the
seedling from the tray. The automatic transplanter requires either plug or pot seedlings.
Removal of the individual seedlings from the trays at rates of 3 to 5 plants s-1 is a
challenging task (Shaw, 1999). The initial investment on production of plug is high and
the mechanisms employed for the removal of seedlings from tray are complex.

In India, past studies reveal that fully automatic vegetable transplanters capable of
feeding and metering individual push type seedling ball removal have not yet been
developed.

The existing system of vegetable cultivation is to transplant the seedling manually

with the help of women labours. The vegetable cultivated area is increasing day by day, and
the vegetable seedling needs to be transplanted within a short period of 25 to 30 days.
Nowaday labour is not available for doing this job due to drudgery involved in this operation.

Several attempts have been made elsewhere in the world and prototype have been
brought out. In one model the seedlings were pushed into a device that rotated them to a
vertical orientation after which they were transported pneumatically to preformed holes
in the ground (Cayton et al., 1981). Another model has a vertical plunger that pushes
plant cell upward out of the tray, where they are grasped by mechanical grippers before
being transported to the ground (Bryant et al., 1984).

5
Objectives

Keeping in view of the above facts, an investigation titled “Development of a

tractor mounted automatic vegetable transplanter” was taken up with the following
specific objectives.

1. To develop a suitable technique of protray system of nursery seedlings to

get accommodated in the contemplated transplanter,

2. To conceptualize and develop the seedling transfer and planting mechanism,

3. To evolve the logistics of removal and transport of plugs to the furrow,

4. To fabricate a prototype automatic vegetable transplanter, and

5. To evaluate the field performance of the vegetable transplanter in terms of

the precision in intra spacing and plant stand/uprightness.

6
Review of Literature
 CHAPTER II

REVIEW OF LITERATURE

In this chapter, the literature pertaining to development evaluation and

performance of vegetable transplanters are presented in the following subtitles.

2.1 Growing and transplanting of vegetable seedlings by conventional methods

2.2 Existing semi automatic vegetable transplanters and their metering

mechanisms.
2.3 Kinds of protray configurations and growth media

2.4 Automatic vegetable transplanters and mechanisms for plant removal

and planting

2.5 Furrow opener and closer mechanisms for vegetable transpalnters

2.6 Seedling and plug characteristics

2.7 Performance of the vegetable transplanters

2.1 Growing and transplanting of vegetable seedlings by conventional methods

Vegetable seeds are conventionally sown in flat levelled field, usually in a

selected corner of the field. The strip is thoroughly ploughed three or four times. The
clods are broken by hand beating to make small particles and levelled for uniform
distribution of irrigation. Farm yard manure is evenly spread over the levelled field.

Extracted seeds are usually dried for two or three day in open sun drying and
then sown. Commercially available treated seed are now popularly used for raising the
nursery. The seeds are broadcast in the field and a thin layer of about two to three mm
high river sand is spread over the field and irrigated. Coconut rache or paddy straw is
spread over the seed to avoid direct sun light to protect the tender seedling.
After germination the coconut rache or paddy straw is removed and seedling irrigated,
25 to 30 days old tomato seedlings are then removed by hand and washed with water
to separate the single seedling for transplanting. The transplanting is done in the main
field manually either on raised bed or flat bed.

7
2.1.1 Conventional method of growing vegetable seedling

Chandra, 2001 stated that for raising the vegetable seedling for one ha, 250 m2
area is required. The beds are covered with a layer of farm yard manure and sand
mixed in equal proposition. Farm yard manure @ 4 kg m-2 was used to protect seeds
and seedlings, the beds were treated with 10 per cent formaldehyde. Before sowing,
the nursery beds were drenched with 0.2 per cent Dithane M - 45 or treated with
Difolation or Captan. The seeds were treated with Thiram or Bavistin @ 2 g per kg
and soon after sowing, the beds were irrigated with rose can and covered with paddy
straw. The beds should be irrigated every day morning. Before transplanting, seedlings
were treated with insecticides 0.1 per cent monocrotophos and mancozeb, 0.2 per cent.
Hardening of seedling before transplanting is essential. Hardening was done by
withholding water 4 - 5 day before uprooting seedling. Adding 4,000 ppm sodium
chloride or spraying of 2,000 ppm CCC was effective for hardening of seedling.

Sachan et al., 2010 conducted an experiment, for raising nursery in a highly

porous and fertile media mixed with well - rotten compost or FYM. Raised beds were
prepared and seeds sown in lines of 50 mm apart covered with sand. For early
germination of seeds, nursery beds were covered with dry paddy straw. He reported
that about 25 to 30 days old healthy and vigorously grown seedlings having 5 leaves
are to be used for planting in the main field. Transplanting was recommend either in flat
bed (in light soils) or on ridges (in heavy soils) with the spacing of 750 x 600 mm in
autumn and winter and a spacing of 750 x 400 mm for spring and summer seasons.

2.1.2 Conventional method of transplanting

Ferminger, 1953 reported that in India, for small - scale vegetable gardening,
holes of 600 mm diameter and 300 mm deep are manually dug in the field at desired
spacings. The soil is mixed with farm yard manure, bone meal, and wood ashes. The
hole is then filled to a depth of 150 - 200 mm and packed. A seedling is placed in the
middle of the hole and topsoil is filled around the seedling, compacted, firmed, and
soaked with water. A shelter is built to shade the seedling under dry weather conditions.

Rotty, 1960 reported that transplanting of vegetable crops can be done by hand
or with a transplant board. The transplant board consists of a board of wood or
aluminium, or other light metals, with channel or angle bar cross section, to which

8
seedlings are attached. A trench is made in the field before transplanting. The
transplant board is transported to the trench and the seedlings are released. The soil
was covered around the seedling using a spade and then tamped. To be efficient, three
boards are needed for each person doing the planting; i.e., one board awaiting filling,
one being filled, and one being planted. The transplant boards had no standard length
and boards as long as 2550 mm accommodating 75 seedlings were used.

Pugh, 1983 found that labour requirement for transplanting tobacco is

approximately 1.12 h for 1000 plants (15 plants min-1).

Boa, 1984 conducted experiments with manually fed transplanters and

concluded that the work rate of 1300 - 1700 plants (operator h-1) were recorded with
cabbages spaced 360 mm apart. Plants which were within 30 of vertical were
considered likely to produce upright plants.

Margolin et al., 1986 found that the average rate of planting per work is limited
to 30 - 40 plants min-1.

Suggs et al., 1989 stated that the labour requirement for pulling seedling from
field beds and planting them with hand fed transplanter are approximately 2.5 h per
1000 plants.

Orzolek, 1996 observed that manual transplanting on a large commercial scale

is labour intensive, expensive, and often does not result in uniform distribution of
plants compared with mechanical transplanters.

Chaudhuri, et al., 1999 carried out a survey to assess mechanization gaps in

field of seedling, planting and transplanting. The results of the survey indicated that there
was a mechanization gap in area of transplanting of vegetable crops. Labour requirement
for manual transplanting for various vegetable crops are shown in table 2.1.

Kim et al., 2001 observed that in Korea, manual transplanting of Chinese

cabbage (B.camperstris L.var. Pekinensis) required 184 man - hours ha-1. Manual
transplanting is labour intensive, expensive, and results in non - uniform distribution
of plants in the field.

Anonymous, 2004 found that raised bed planting requires 185 - 260 man - hours
ha-1 for transplanting egg plant, onion (Allium cepa L.), and chilli pepper (C. annuum L).

9
 Anonymous, 2004 concluded that flat planting method requires about
160 man - hours ha-1 for transplanting tomato at 600 mm row - to - row spacing and
450 mm plant - to - plant spacing.

Table 2.1 Labour requirement for transplanting of selected vegetable crops

Plant geometry
Labour requirements
Crop ( row x plant spacing)
man - day ha-1
in mm
900 x 450 15
Tomato 600 x 450 20
450 x 450 25
600 x 450 20
Chillies
450 x 450 25
75 x 75 100
Onion 125 x 125 50
150 x 100 50
Cabbage 450 x 450 30
Cauliflower 600 x 450 30

Ghai and Arora, 2007 concluded that most of the vegetables like cucurbits
(Cucurbita spp.), beans (Phaseolus spp.), okra (Abelmoschus esculentus) and leafy
vegetables are sown directly in the field. Vegetables like tomato (Solanum
lycopersicum), eggplant (Solanum melongena) and peppers (Capsicum spp.) are first
sown in nursery beds and later transplanted manually either on ridges or on a well
prepared seedbed.

Kumar and Raheman, 2008 found that manual transplanting of seedlings is

labour intensive, expensive, time consuming and often results in non - uniform plant
distribution.

Singh and Singh, 2009 conducted a study in vegetable producing districts in

Madhya Pradesh to evaluate the economics and assess mechanization gaps of
vegetable cultivation. The study revealed that there is a scope for mechanization of
selected operation i.e, transplanting/planting. Human labour utilization was observed
at a maximum of 277 man - day ha-1 in garlic cultivation followed by tomato 271 man
- day ha-1, Onion 200 man - day ha-1, cauliflower 168 man - day ha-1 and brinjal

10
138 man - day ha-1. Transplanting of brinjal, tomato, cauliflower and cabbage were
done manually, which required 15 - 27 man - day ha-1. He reported that the vegetable
growers needed tractor operated transplanters, which maintain a 450 x 450 mm or
600 x 600 mm spacing.

Kumar and Raheman, 2011 compared bare root tomato seedlings and paper pot
seedlings. The seedlings were transplanted at 450 x 450 mm spacing after 35 days in
two plots using pot seedlings and bare - root seedlings manual transplanting. 35 days
old bare - root seedlings were used. This required 229 and 320 man - h ha-1 for paper
pot and bare root seedlings respectively.

2.2 Existing semi automatic vegetable transplanters and their metering mechanisms

Rotty, 1960 developed a pocket type planting unit for semi automatic
transplanter for planting bare root seedlings. It had 6 to 12 spring loaded plant pockets
arranged at equal intervals on a drum. The drum was driven from the soil compacting
wheels through a gear or chain drive. When the drum rotate, a plant pocket opens as it
approaches the top, receives the seedling, closes, carries the seedling downwards,
opens, and deposits the seedling in the furrow.

Dooley, 1982 found that one operator per row could only feed 2400
seedlings h-1 with no skips and 2550 seedlings h-1 with one per cent skips.

Burgoyne, 1987 reported the performance of semi automatic machines compare

to conventional method and presented in table 2.2.

Table 2.2 Field planting rates with semi automatic transplanters

Rate per row

Operators
Machine type Plant type
per row Plants min.
Plants min-1
operator -1

2 60.9 30.5
Conventional Bare root tray
2 51.6 25.8

2 64.7 32.4
M - 1000 Bare root tray 2 63.8 31.9
1 35.4 35.4

M - 4000 Tray 1 69.8 69.8

11
 Anonymous, 1988 developed a mechanical model 1000 (M - 4000) transplanter
very similar to conventional semi automatic transplanters except that the chain
carrying the plant grippers is driven by a floatation wheel rather than the packing
wheels and 12 plant grippers are supplied instead of 10.

Anonymous, 1988 developed a mechanical model 4000 (M - 4000)

transplanter specifically for tray - grown plants and has four conical cups per row into
which the operator places the plants. The four cups turn as a unit on a vertical shaft
geared to the ground speed. The cups open at the appropriate spacing to drop the
plants into a delivery tube leading to a furrow opener. There a gate device pushes the
plant into the open furrow.

Hergert et al., 1988 developed a semi automatic transplanter which has either
gripper pockets or flexible discs attached to a conveyor chain to carry transplants from
the loading operator’s station to the furrow. The grippers open automatically to
receive the plant, close to grip the plant and open again at the furrow. The chains
carrying the grippers are driven from the packing wheels with adjustments available
to vary the plant spacing. The flexible steel discs, marked to show where to insert the
plant, carry the plants to the furrow. These machines require two people per row for
loading the grippers or discs, to achieve an acceptable rate. Both types are easily
converted to handle tray plants by adding extensions to each gripper or disc to support
the root media block.

Pelletier and Hergert, 1988 evaluated a Regero HD B4T transplanter with

tomato transplants. Transplants were loaded by hand onto tapered trays or pans
attached to inclined roller chains. As the chain for each row travelled over the lower
sprocket, the pans tipped, dumping the plant root first into the planting shoe cup. With
the transplant in the cup, a cable retracted the rear door as a cam actuated rod push the
cup rearward to seat the plant into the furrow just ahead of the packing wheels.
A scraper at the front of the cup pushed the plant along with any soil debris out the
rear of the furrow opener. The mechanism then returned to the neutral position, ready
to receive the next plant. The planting mechanism, including the pan conveyor, was
chain driven by the ground roller, with provision for adjustment, to achieve timing of
the components and the desired spacing.

12
 Huber, 1991 modified the pocket type planting unit with the use of rubber
grippers for holding the seedlings in a machine using a two jaw seedling holder which
discharges the seedlings into the furrow.

Brewer, 1997 devised the operation of a Model - 5000 transplanter. The seedlings
were paced in cups, six of which were arranged in a circular array. The array rotated so
that the cups pass sequentially over a vertically aligned tube. The seedlings were dropped
into tube, pass through the tube and land in a V shaped shoe which held the seedling until
the kicker pushed them out into the furrow. Seedlings dropped under the force of gravity,
with minimal interference from brushing against vertical walls. The cups must rotate
68  for seedling to drop completely clear of them, and also increased the drop height for
the Model - 5000 from 250 mm to 640 mm. A seedling was dropped from this height
before it is planted. It takes 0.36 s to travel this distance. During this time interval, the
transplanter moved forward by 150 mm. If a transplanter was hand fed, which implied
planting rates under 2 seedlings s-1, than drop height down to 380 mm should be
acceptable. The Model -5000 mechanical transplanter was modified such that the number
of cups (into which seedlings were placed) increased from four (EM 4000) to six (5000).
The general shape of cup and the mechanism to release was also modified. These
modifications made it easier and faster to load and unload seedlings. A kicker which
pushed seedlings out of a shoe into the soil furrow was added and the kicker drive
mechanism was changed from continuous to intermittent. Hand feeding rates were limited
to about 60 to 90 seedlings row -1 minute-1.

Feng et al., 2000 developed a conveyor type planting unit, which has a
horizontal belt conveyor driven from a soil compacting wheel shaft or ground wheel
shaft. The belt may be either plain or have a series of cross - wire partitions
(compartments, Margolin et al., 1986), on which seedlings are placed. The conveyor
delivers a seedling to split cone cups in a mulch planter or transplanting disc in a
regular automatic transplanter by setting it in an upright position.

Kim et al., 2001 developed a vertical descending cup or bucket - type planting
unit for semi automatic transplanter. Here a conical cup was provided on a drum,
which was driven with a soil compacting wheel. The cup maintained a vertical
orientation throughout the rotation, so that the seedling fed to it was carried vertically

13
to the furrow before planting. Seedlings fall by gravity into the furrow as the cup is
lowered.

Anonymous, 2002 concluded that the imported vegetable transplanters are quite
sophisticated giving higher performance and the semi automatic vegetable
transplanters developed earlier was found to have poor performance on traditional
seedling because it had a gravity type metering mechanism.

Chaudhuri et al., 2002 reported that two types of mechanical transplanters were
imported and modified to use under Indian conditions. The feeding of seedlings was
carried out manually by persons sitting on machine. One person was required for each
row. Finger type and cone type metering mechanism were used in these transplanters.
The tractor operated 6 row and a self propelled one row vegetable transplanter were
evaluated for planting brinjal and tomato crops. The average seedling height during
transplanting ranged 160 - 250 mm. Field capacity of the transplanter ranged from
0.08 to 0.10 ha h-1 depending upon row and plant spacing. The labour requirement was
about 20 - 25 man - h ha-1 as against 300 - 320 man - h ha-1 in manual transplanting.

Garg and Dixit, 2002 developed a single row semi automatic transplanter, that
plants bare root chilly seedlings. It had a single cone type metering mechanism with a
drop chute for placing seedlings into a furrow by gravity. Two operators alternately
place a single seedling at one time. The rotating plate strikes the cone opening it, and
the seedling moves in the drop chute pipe.

Anonymous, 2004 reported the field performance of a two row tractor mounted
semi automatic transplanter with pocket type planting unit. The row spacing for
tomato, cabbage, and cauliflower were 670 mm and plant spacing 250 to 300 mm; and
chile pepper row spacing was 670 mm and plant spacing 500 - 540 mm. Suitable
forward speed of operation for obtaining a minimum of missed planting was found to
be 0.9 to 1.1 km h-1 for various crops.

Anonymous, 2004 reported the optimum forward speed as 0.9 km h-1 and field
capacity as 0.1 ha h-1 for planting tomato at 600 mm row spacing and 450 mm in row
plant spacing using tractor drawn two row semi automatic transplanter with drum type
planting unit.

14
 Craciun and Balan, 2005 developed a rotary cup type planting unit. This type of
unit has a shaft rotating in a horizontal plane, to which 6 - 8 cups are attached in a
circular fashion. The drive shaft was driven by the soil compacting wheel shaft. As
the shaft rotates, cups are fed with seedlings, which were then carried to the bottom of
the cup and dropped in the furrow. In another arrangement, cup bottoms were open
and supported on a horizontal stationary plate that had an opening, into which the
seedling was discharged.

Parish, 2005 designed a rotary cup type planting unit, which allowed the
operator to place several seedlings rapidly and then have a brief time to untangle
seedlings or remove seedlings from cells rather than having to maintain exact timing
for each seedling. It was used for planting pot seedlings in a semi automatic vegetable
transplanter. The transplanter with this type of unit planted 50 - 80 seedlings min-1
row -1 depending on the row spacing. Each planting unit required only one person for
placing seedling in cups.

Anonymous, 2006 developed a tractor mounted 2 row and 3 row semi automatic
vegetable transplanter in India for bare root seedlings and plugs. Pocket type metering
devices have been provided in the transplanter for bare root seedlings. The field
capacity and labour requirement have been reported to be 0.082 - 0.092 ha h-1 and
44.4 man - h ha-1 respectively at a forward speed of 0.8 - 1.0 km h-1. Rotary drum type
metering devices have been provided in the transplanter for plugs. The field capacity
was reported as 0.14 ha h-1 and labour requirement as 28.6 man - h ha-1, when
operated at a forward speed of 1.4 km h-1.

Satpathy and Garg, 2008 used finger type metering mechanism for holding and
placing the seedling. The picking finger had spring mounted rubber flaps, which
opened before passing through the tunnel and closed during its passage. Again the
flaps opened at the bottom end of the tunnel to release the seedling in a furrow.

Manes et al., 2010 developed a tractor mounted two row semi automatic
vegetable transplanter. The picker fork had a spring mounted rubber flap, which
closes while passing through the tunnel. Again these flaps open at the bottom of the
tunnel to release the seedling in a furrow. The inclined wheel compact the soil around
the seedling.

15
 Narang et al., 2011 developed a two row vegetable transplanter with
revolving magazine type metering mechanism. The height of the revolving magazine
type metering was selected on the basis of height of vegetable seedling crops. Cell
diameter of 75 mm and cell height of 135 mm was selected. The seedlings of brinjal
crop were grown in 98 cell plug tray in soilless media. Nine holes, each of 75 mm in
diameter were arranged on circular periphery of a metal plate of 340 mm diameter. Nine
pipes, each of 70 mm length, were welded on the holes in the metal plate. A 100 mm
diameter drop pipe was used for easy falling of seedling during transplanting. The
designed drop chute pipe was such that the plants falling at the ground make an angle
of 85 to 88  with the horizontal, and falls in the direction of travel of the machine. To
make plant upright, the drop chute pipe was inclined by an angle of 15 to vertical.

2.3 Kinds of protray configuration and growth media

2.3.1 Different kinds of seedling plugs

Nambu and Tanimura, 1992 developed an automatic transplanter using a chain

pot for vegetable crops. The chain pot (paper pot) is made from two ply sheets of
paper joined together with water resistant adhesive to form individual paper pots.
Each pot is connected by a paper joint, which forms a chain of paper pots. Each joint
has a perforated section which facilitates separation during transplanting. These joints can
tolerate regular forces pulling in a parallel direction, but can be easily torn by oblique
tensile forces, with no damage to the paper pots or seedlings. Plants were grown in the
chain pot in 25 x 38 mm (diameter x height), 29 x 38 mm, and 35 x 50 mm cells.

Brewer, 1995 evaluated the performance of an experimental automatic seedling

transplanter with static cassettes. The pepper seedlings used for this study were grown
in Styrofoam trays. Sturdy seedlings were ready for transplanting at 4 to 6 weeks after
the seeds were planted. The height of plug and seedling top was 233 mm, plug height
75 mm, weight of plug and seedling top was 17.8 g and plug weight 16 g.

Tsuga, 2000 developed and standardized cell mold seedling trays and pulp mold
cell trays (paper pots) for possible use in a fully automatic vegetable transplanter. The
cell mold seedling tray was made of plastic and there were three types of trays,
containing 128, 200, or 288 cells, in which the arrangement of cells was 8 x 16,
10 x 20, and 12 x 24. The shape of the trays was that of an inverted pyramid. The pulp

16
mold cell pot was a kind of paper pot, in which the pulp mold was made of used
paper. The pot dimensions were 30 mm2 (128 pots) and 25 mm2 (200 pots) and the
seedlings were arranged in the form of a pyramid. The arrangement of cells was 8 x 8
and 10 x 10 and each pair of pots was arranged on a under tray of cells. The diameter
of seedling hole for paper pulp mold pot was18 mm for 128 cell and 15 mm for 200
cells and for cell tray 12 mm, 9 mm and 7 mm for 128, 200 and 288 cells respectively.

2.3.2 Effect of different growing media

Charles and Mark, 1990 used plug trays filled with vermiculate. Trays were
280 x 550 mm and can hold 200, 406 or 648 plants per tray with cell volume of 90, 30
and 13 mm3 respectively. Water was misted for 5 s every 6 to 12 min during day light.
The plastic cell packs filled with a medium of 1, loam soil: 2, peat: 2, perlite (by volume).

Cheong et al., 1992 reported that fermented hog manure - sawdust and cattle
manure could substitute for peat moss in potting media.

Huang and Ai, 1992 concluded that in the conventional tray cell the plant roots
spiral around the cell bottom resulting in retarded root branching and root
tangling/root binding. The upper portion of growth media was not well utilized and
some roots tend to grow upward. He used growing medium of peat - soil mix.

Handreck and Black, 1994 found that potting mixes with more than
30 per cent soil by volume usually have poor aeration in pots. It was observed that
these mixes also have a high bulk density and can have a low level of available water
if too much clay in the soil. Soil contains pathogens including weed seeds, which are
normally destroyed by air - steaming. Used sand is sometimes included as an
ingredient in growing media substrates. Those with medium to very coarse particle
sizes were generally preferred as are sharp sands since rounded particles can separate
out during mixing. Sand provided ballast and helps overcome re - wetting problems.

Menzies and Aitken, 1996 found that 10 per cent of peat can be substituted with
fly ash for vegetable transplants.

Subler et al., 1998 observed that the best tomato seedling weight occurred in
plug trays when vermicomposted pig manure constituted 10 to 20 per cent of the total
volume of the medium.

17
 Atiyeh et al., 2000 observed that the vermicompost added to commercial potting
mix improved seed germination, enhanced seedling growth and development, thus
increased overall plant productivity of tomato.

Press, 2001 stated that the bare root seedlings were obtained by pulling
seedlings directly from nursery beds. The soil block seedlings were obtained by
sowing of seeds in moist soil cubes made by mixing soil, peat, compost, and sand.

Arenas et al., 2002 found that media with more than 50 per cent coir reduced
growth of tomato plants compared to peat - grown control plants.

Kuepper and Everett, 2004 recommended a mix of 3 parts coir to one of

compost and another with a product containing 35 to 45 per cent peat moss,
vermiculite and pine bark.

Paul and Metzger, 2005 reported improvement in quality of eggplant and pepper
transplants and reduction in quality of tomato transplants when vermicomposted cattle
manure was substituted at rates up to 20 per cent.

Colla et al., 2007 concluded that the coir pith had a better balance between
water and air capacity and can be used systematically in all areas of growing media
production. They recommended it to be used as a peat replacement or in a reduced
peat mixture. Trials were conducted for growing lettuce transplants in coir alone and
peat/coir mixes and found that the optimum combination was about 50 per cent peat
and 50 per cent coir.

Schofield, 2007 used soil - based growing media which was the norm prior to
the advent of soilless media based on peat. A ‘base mix’ to be added to a growers
own soil for propagation purposes and then two loam - based growing media for seed
raising were used. He concluded that the major problem with soil was that of
maintaining access to a supply of consistent quality.

Singh et al., 2007 reported that the quality of tomato transplants was highest for
transplants grown in round cells with a volume of 68.2 cc.

Schmilewski, 2008 concluded that the chemical and physical characteristics of

coir materials vary greatly with their origin, time in storage and the duration of the
treatment process.

18
 Kumar et al., 2009 prepared growing media for starting seedlings which
contained peat and vermiculate plus small quantity of plant nutrients. Also used a
mixture of two parts garden soil, one part peat or coir fibre, and one part medium to
coarse sand (2 soil:1 coir : 1 coarse sand) and suggested that if the garden soil is high
in clay to use a 1:1:1 mixture of these materials.

Kumar and Raheman, 2010 conducted an investigation to identify the mix and
volume of vermicompost based potting mixes required for production of good quality
transplants of tomato (Solanum lycopersicum L.), eggplant (S. melongena L.), and
chili pepper (Capsicum frutenscens L.) in paper pots. Paper pots of 65 cc in volume
filled with a mix of 25 per cent vermicompost, 100 cc in volume filled with a mix of
50 per cent vermicompost, and 81 cc in volume filled with a mix of 33 per cent
vermicompost, in combination with equal parts of sand and top soil were most
suitable for production of good quality transplants of tomato, eggplant, and chili
pepper, respectively. It was observed that sand adds air space and weight to the
potting mix. Clean, washed sand has a near neutral pH and little if any nutritional
value for plants. Tomato had a recommended seedling stage of 3 to 4 weeks and the
eggplant and pepper have a recommended seedling stage of 4 to 5 weeks.

2.3.3 Effect of cell size of protray

Anonymous, 1983 developed a plant seedling production and transplanting

system with a unique stabilized peat growing module. These modules were handled in
a plastic cell tray which was used both in the green house and for field handling. The
individual modules were made up of finely ground peat bound together with a special
polymer and were formed with a cavity to receive the seed. A fully automatic
transplanting machine was developed that could remove the modular plants from the
trays and set them into the soil at the rate of 30,000 plants h-1 for a 4 row machine for
lettuce and celery seedlings.

Greer and Adam, 2005 found that the cell size affected the watering schedule.
For larger containers, water must be added to thoroughly moisten the entire medium
profile, whereas for smaller containers a less than saturating amount of water can be
added without detrimental effects to roots since the water will distribute adequately.

19
 Anonymous, 2008 concluded that the choice of cell size had a clear effect on the
growth of the organic transplants and was more important than the choice of growing
medium. Transplants grown in the larger cell sizes, provided individual plants with a
larger volume of substrate, tended to be larger and of superior quality.

2.4 Automatic vegetable transplanter and mechanisms of plant removal

2.4.1 Plug plant removal mechanism

Huang, 1973 developed an automatic transplanter for tobacco, which used

truncated pyramid soil blocks held in plastic trays. Automatic transplanting was
accomplished by pulling the blocks out through the bottom of the growing tray with
vacuum and allowing them to drop down into a soil furrow. The plants were not
carried directly to the soil, but had to depend on the parachute effect of the foliage to
keep them upright.

Cayton et al., 1981 designed a plant removal from horizontal pushing from a
vertical tray. This type of removal techniques handled eight plant modules at a time
that were pushed into a device that rotated them to a vertical orientation after which
they were transported pneumatically to preformed holes in the ground.

Moden and Hauser, 1982 devised three cutting devices namely house hold
scissors, solenoid operated surgical knives and hotwires to severe the polyethylene
between the plant cells so as to drop them. Among the three scissors was the best.

Armstrong and Hanacek, 1984 developed a plant removal mechanism wherein

the plants were pushed onto a conveyor with special holding flights delivering them to
a plant setting mechanism. The seedlings tended to stay in an orderly queue but the
plants had to be reoriented for planting. It was possible to manually drop plants into a
horizontal queue to gap up the string, while it was being transferred to the plant
setting mechanism.

Bryant et al., 1984 developed an upward removal from a horizontal tray. It had
a vertical plunger that pushed plant cell upward out of a web. These plungers pushed
individual seedling upward to where they could be grasped by mechanical grippers
before being transported to the ground.

Munilla, 1984 developed a peat block transplanter that cut squares of compressed
peat containing seedling plants and transferred them to the ground. This system was not
20
successful because the peat blocks were not dimensionally stable and changed in length
and width with variations in moisture content. The separated blocks frequently did not
contain seedling plants and the cutting mechanism damaged the plants.

Anonymous, 1985 developed an automatic transplanter for planting bare - root rice
transplants. These utilized mechanical fingers that grabbed plants from a growing mat of
seedlings and set them into puddled soil. One or several plants are planted in each bunch.

Anonymous, 1985 developed a paper chain of plant units for sugar beets.
Honeycomb shaped paper cells were formed with both water soluble and water proof
glues. As plants were produced in the peat filled cells, the joints between the cells
made with the water soluble glue came apart resulted chains of seedlings or seedling
modules. These can be separate and set the modules to an automatic transplanter and
drop the plant modules into a furrow or positively place them to the soil and hold
them while the soil is being formed around the plant roots. The paper cells deteriorate
in the soil as the plant developed.

Branch, 1985 developed a fully automatic transplanter for lettuce and celery
transplanting. It utilized a bottomless plant cell tray that made possible automatic
setting of plants by pushing them through the bottom of the tray into the soil. The
seeds were placed off - centre in each cell so that the seedlings develop off - centre
and are not damaged by the plunger that remove them from the cell. The capacity of
an 8 row transplanting machine was reported to be 33,000 plants h-1.

Branch, 1986 developed a bottom removal technique for removing seedling

from a tray. A spiked plunger was utilized to push individual seedlings out through
the bottom of a rectangular tray and set them directly into moistened soil.

Margolin et al., 1986 developed a belt conveyor type planting unit suitable for
pot seedlings in automatic transplanters. The seedling pick - up unit extracts an array
of seedlings and directly feeds onto the conveyor. One advantage of this type of unit
is that a substandard seedling can be easily identified using a suitable vision system
and replaced with good quality seedlings as the seedlings are carried by the conveyor.
It can transplant 84 - 90 seedlings min-1 row -1 .

Suggs et al., 1987 designed a self feeding transplanter with a roll - feed
mechanism, where the line of cells were passed between a pair of counter - rotating

21
rollers which fed a line of Z folded plants into a pair of acceleration rollers. These
rollers had a peripheral speed several times as great as the feed rollers. The rollers tore
the plant from the strand and propelled them into the drop chute. Plants fall by gravity
into the furrow just before it is closed by the press wheels.

Feldman et al., 1988 evaluated a castle and cooke transplanter for tomato
transplanting. Planting units were set up in two row tandem offset sets of 4, to provide
8 rows. One tray rack was provided for each row set so that one operator installs full
trays and retrieves empties for four rows. Plants were ejected from the trays one row
at a time by plungers that engage the root end of the plug through holes in the tray
bottom below each cell. Some trays were grown at one plant in each cell and the
remainder at one plant every second cell, with the seedlings ready to plant.

DeGroot, 1989 developed a seedling removal mechanism for high speed

transplanter. Whole rows of seedling were mechanically pushed from a vertical tray as
they were intercepted by individual plant grippers that transfer and distribute them
into a furrow. This transplanter had high capacity because whole rows of plants are
removed at one time.

Rumsey et al., 1989 designed a two row electro - hydraulic vegetable seedling
block transplanter. It was designed to automatically remove cylindrical seedling block
51 x 51 mm from growing trays and feed them to conveyor type chutes. The depotting
and feeding mechanisms consisted of a tray conveyor, indexing device, and extractor.
The tray conveyor provided unidirectional movement of the seedling tray. The
indexing device aligned two rows of four seedlings to the proper extracting position.
The extracting device was designed to depot two rows of four seedlings from the tray
during each operating cycle. The feeding mechanism was used to both guide the
seedling and push them onto a chute. The guides were two pairs of parallel metal
plates. The two pushes were made from concave aluminium blocks attached to
pushing shafts. The guides and the pusher were independently powered by separate
hydraulic cylinder.

Lloyd Arnold, 1990 developed a clay - mill type automatic transplanter wherein
two tray holders held plastic trays in a horizontal plane facing each other. Extraction
of plug from the 288 cell trays by a common set of grippers mounted on a slide
between the two tray holders. Mechanism was designed to reduce the problems of root
22
binding between the cells of plastic trays. Two plant indexing mechanisms which held
plugs in proper alignment for delivery to shoe and plant in a furrow. He also developed a
Lannen automatic transplanter for planting tomato plug plants from special plastic trays
(open bottom type) of two different densities (144 cell and 256 cel1). This unit was not
equipped with any compensatory mechanism for missing cells. Labour is required
only to load the trays into tray holders. The lannen automatic transplanter employed a
plunger to push the plug out from the bottom of the tray and onto a pair of rigid
needles. The needle pair holding the transplant then rotated 90  from horizontal to
vertical where another pair of adjustable rubber covered fingers wiped off the plant
into the delivery tube to the planting furrow. Final transplant was gravity dependent.
The planter was evaluated with tomato seedling removal gripper mechanism. The
inability to grasp and extract all the plants reduced initial plant stand significantly.
Slippage of the most seedling stem while in the grasp of the gripper fingers can result
in non - delivery of plug to the plant indexing mechanism. Positive grasping
capabilities of the steel grippers were greatly enhanced with the application of
rubberized coating material that increased plant extraction rate from trays and enabled
better alignment of plugs into plant indexing mechanism. A hydraulic orbit motor
controlling the rate at which the plugs were pushed from tray on to needles and then
wiped off these needles down the plant delivery tube. Once the hydraulic flow rate
was set, field plant spacing became a function of forward ground speed (slower speed
giving closer plant spacing).

Lloyd Arnold, 1990 found that the cylindrical shape of the stem, by nature,
appeared to be capable of withstanding varying levels of externally applied forces
until the extreme limit at which the outer cell membrane is broken or the stem is
crushed. These levels appeared to far exceed the removal force required when pulling
plants from plastic trays. The occurrence of root binding between cells increased the
resistance to extraction and plant slippage in the grasp of the grippers. Hence instead
of gripper air propulsion technology was used to propel plugs down to the furrow and
reduce gravity dependency when planting tall plants. Tall plant heights required
higher air pressures. Foliar entanglement of plants in the indexing mechanism was the
main reason for increasing line pressure to the venture. The dangling root hairs
required higher rates of air delivery.

23
 Williams, 1990 developed an upward removal from a horizontal tray, where a
vertical plunger operated through the cell drain hole.

Cooley, 1991 designed a plant removal mechanism where the horizontal

removal of seedlings from a vertical tray was done by way of a blast of air that blows
them into a string of cups which in turn delivered the seedling to individual plant
setting mechanism. Each plant must reorient itself from a horizontal to an erect
orientation while it is falling into a cup. It was concluded that control of plant
trajectory was difficult with this system especially if the foliage is entangled.

Huang and Ai, 1992 developed an air pruned transplant production system for
fully automated transplanting. It performed pulling, singulating and transplanting in
one simple operation. The air pruning tray becomes an integral component of the fully
automatic transplanter in which the individual seedling is pneumatically removed
downward through the tray cell after removal of the detachable bottom. The vacuum
system removed the seedling from the open bottom of tray cells, and as the suction
pulled the seedling downward, the leaves fold aerodynamically so that no injuries
occur. The downward moving seedling can achieve a velocity in excess of 6.1 m s -1.
Two DC motors index the frame and trays in X and Y directions to position a cell
with an air - pruned seedling over each suction drop tube and the impulse vacuum
system instantly removed the seedling from the tray cell, transplanted into flat basket
or field. A self contained computer controlled all operations for accurate positioning
of tray. Approximately 85 seedlings min-1 were automatically transplanted by each
suction drop unit. Empty tray removal and full tray addition to the transplanter while
sliding off the tray bottom was quickly accomplished, and the indexing mechanism
resumed operation immediately. The electric eyes sensed an approaching pot,
computed its speed and time of seedling removal from the tray cell to reach the pot as
it arrived under the drop tube.

Suggs et al., 1992 developed an automatic feeding system for corn and tobacco
plants. It utilized a cell tray having 10 cells per row and 2 rows per tray. Cells were
pyramidal in shape with a square base of 25 x 25 mm and height of 70 mm. Plants
were clipped three to five times to remove excess foliage and were transplanted when
they were approximately 150 mm tall. The plants were fed 10 at a time directly from
the tray by a set of push rods, which entered through holes in the bottom of each cell.
24
The push rods forced the plant root balls onto a set of small rods having the same
centre - to - centre spacing as in the cells. The rods were mounted on a vertical roller
chain which moved the plants downward and into the furrow just in front of the press
wheels which closed the furrow on the plant roots causing the plants to be pulled off
the rods. As the roller chain passed around the radius of the lower guide, the tips of
the rod spread out to space the plants and to provide a rear ward motion
approximately equal to the forward motion of the transplanter. This velocity
compensation allowed the plant to approach the ground with virtually zero horizontal
velocity so the furrow can be closed on its roots before it is removed from the rod.
Thus, its position is controlled at all times. The rods and roller chain moved
continuously and it was necessary to move the tray down in synchrony with the rods,
while the plants are being transferred to the rods. When the transfer was complete, the
tray returned to its initial position. At the very beginning of the downward motion, a lug
engaged a cross fed system which fed the tray laterally to line up the next row of plants
with the push rods. A metal perimeter having slots aligned with the tray rows was placed
around the tray to provide positive alignment of the tray with the push rods.

Faulring, 1993 designed a plant removal mechanism, which pushed horizontal

seedlings from a vertical tray with plungers operating through the drain holes and to
impale them on pins that support them until they can be reoriented and set. The
individual plants were removed, impaled and rotated to an upright orientation before
they were dropped into plant setting mechanism.

Kohli, 1993 optimally designed a four bar mechanism for a 10 row paddy
transplanter for washed root seedling. The fingers opened 10 mm when penetrating
into the seedling stalk (tray) and the depth of penetration of fingers into the puddled
soil was 20 mm. The fingers opened by 15 mm or more when releasing seedling in the
soil. He observed that the maximum force acting on finger tips is 0.07 N.

Tesch and Bierman, 1993 developed a plant removal mechanism for the
preparation of pot seedlings. Here, the seedling pick - up unit had an upper plate with
four pinholes spaced at the corners of a square of a determined width. A lower plate is
adjustably spaced beneath the upper plate and had four pinholes located at the corners
of a square of a lesser width than the upper plate. Four transplant pins extend between
the upper and lower holes so that when the upper plate is brought close to the lower
25
plate by a pneumatically activated mechanical linkage, the pins converged and hold
the seedling plug. A pneumatic actuator was connected to the lower plate, which lifted the
pick - up unit with the seedling plug. The mechanical linkage was operated to move the
upper plate away from the lower plate, so that the pins diverge and drop the seedling.

Brewer, 1994 designed a conceptual model of an automated seedling transfer

from growing trays to shipping modules for tomato and pepper seedlings. The
functions of the experimental model contained the essentials of a machine, such as
ejecting plugs from cells, gripping stems of seedlings and lifting them, transferring
seedling from tray to module, and dropping seedling into a module. Packers extract
seedling from growing tray, scan seedling for substandard plants (culls), discard culls,
and drop seedlings into modules. The grippers array is stationary in the horizontal
plane. Modules, trays, funnels, ejectors, and shields move with respect to tray. Two
hundred grippers are to be arranged in 10 rows of 20 grippers each. The distance
between grippers within a row is the same as the distance between the seedlings in
tray rows (38 mm). A tray moves so that stems of seedlings in one row are within the
grippers in an array row. Seedlings are to be extracted by ejectors giving a short blast
of air under the seedling plugs, grippers closing on the stem, and cylinders lifting the
row of seedling out of the tray to a height so that bottoms of plugs will clear the tops
of seedling left in the tray. The tray is to move forward (to the right) so that its next
row of seedlings are within the grippers in the next row of the array, and the process
repeats. Stepping motors control movements of the trays, modules, funnels, and
ejector on the primary packer.

Brewer, 1995 designed a dual cassette transplanter which had an automatic

seedling feeder. Seedlings were presented to the transplanter already singulated and
graded in shipping modules. Seedlings dropped by gravity from the shipping modules
to the cassettes then to the dibbles that planted the seedling. The shipping modules
were moved under computer control to feed the cassettes, but the modules were
‘captive’. That is, they could move forward or backward only through a set distance.
A magnetic reed switch was used to detect magnets that were attached to each of the
dibbles. As the ground wheel driven dibble wheel rotated, it activated the reed switch,
which was monitored by the computer. The computer controlled the cassettes and
shipping modules such that a single seedling was dropped into each dibble.

26
 Brewer, 1997 developed a cam operated kicker, which allowed a period of time
during which a seedling can enter the shoe without interference from the kicker.

Shaw, 1997 developed a plant removal mechanism for an automatic vegetable

transplanter. The plant removal mechanism received tray in a vertical orientation with
the plants horizontal. The seedlings were sequentially ejected from the tray cells of
each row and the trays advanced downward as each row is emptied. The seedling
plants were removed by pneumatically operated ejectors acting through the cell drain
holes and each plant fell between flights of a horizontal conveyor which delivered
them to the plant setting mechanism. With many seedlings, there was interference
from the seedling in the row above the row being emptied. So a moving plant divider
was developed that lifts the foliage of seedling above each plant as it is removed, thus
reducing the tangling of foliage between the seedlings.

Williames, 1997 developed indexing drum type seedling pick - up unit. This
type of seedling pick - up unit was used for picking up seedlings from the tray by
guiding the tray to the position, from where seedlings were accessed by the pick - up
unit. It consisted of a special tray, indexing drum, and loading frame. The special tray
had a series of indexing grooves and a transverse alignment groove. The indexing
drum consisted of indexing rods, support rings, and an internally mounted plug ejector
subassembly. The indexing grooves and a transverse alignment groove of the tray
engaged with the indexing rods and support rings of the indexing drum. The indexing
drum was mounted adjacent to a loading frame for guiding the plant trays into
position for pick - up of seedlings by the plug ejector. The plug ejector moved relative
to the indexing drum at each of a number of successive indexing positions of the drum
for removal of seedlings from the plant tray. The indexing grooves and transverse
alignment groove on the plant tray provided datum surfaces for ensuring accurate
positioning of the plug ejector subassembly relative to the seedlings in the plant tray.

Shaw, 1999 stated that the first mechanism developed for seedling removal was
a pegged board that had wooden dowels arranged so that they could register with the
tray drain holes and would push all the seedling out of the tray when the tray was
pushed down over the pegs. Another plant removal mechanism instead of pegs
developed subsequently used combination of mechanical plunger and air jet ejection
for plant removal. It was inferred that the mechanical plungers operating through the
27
cell drain holes effectively pushed seedling out of tray cells under most conditions.
But if the seedling roots are not well developed throughout the module or the roots
have grown into the polystyrene tray material, the root modules may be impaled by
the plungers and not be regularly ejected. The tray grown plants cost more but their
survivability is better and transporting shock is much less.

Tsuga, 2000 designed a cell mold seedling pull out from the tray using a
seedling take out claw of the holding type and discharged them into a bill type opener.
The opener, which is interlocked with the claw, transplanted the seedlings to the
tamped surface of the ridge. He also designed another type seedling removal
mechanism for pulp mold cell pot seedlings. The transplanting claw consisted of
2 blades like claws and an extraction arm. The blade like claws scraped each seedling
pot to separate it from the others and hold it. The arm extruded the seedling pot into
the soil to plant it. Soil was pressed then from both sides of the planted seedling by the
tampering rings to complete transplanting. He revealed some considerations of use in an
array type automatic vegetable transplanting machine. If paper pots were too small, they
were difficult to keep upright on the flat trays in the transplanter. Paper pots above 100 cc
in volume have weak wall strength and are difficult to handle and transport.

Choi, 2001 developed and evaluated a seedling pick - up device for vegetable
transplanters. The pick - up device extracted seedling from a 200 cell tray of seedlings
and transferred them to the place where they are to be transplanted into the soil. The
device consisted of a path generator, pick - up pins and a pin driver. The path
generator was a five - bar mechanism, which comprised of a fixed link, a driving link,
a driven link, a connecting link and a slider. The slider was constrained to move along
the driven link and a fixed slot of combined straight - line and circular paths. The
connecting link joined the driving link and the slider. When the slider moved along
the straight - line path of the slot, it took the seedlings out from the cell and
transferred them to the transplanting hopper, when moving along the circular path.
The device extracted 30 seedlings min -1.

Kim et al., 2001 developed a pin type seedling pick - up unit. The unit had a
pair of pins (claws), which pulled out the seedling one at a time from the tray and
discharged it into the specified spot (furrow, belt conveyor, and tray). The pins were
designed to hold the root or leaves of the seedling of leafy vegetables. The movement
28
of pick - up pins from the point of picking up of the seedling to the point of its
discharge and back to retraction required certain mechanisms. A slider and fixed slot
type of mechanism was used to pick - up the root portion of seedling from the tray and
discharge it into the moving conical type planting unit.

Ryu et al., 2001 used a pneumatic seedling pick - up unit. The fingers were
initially rotated according to leaf direction of the seedling. The fingers were lowered
down and made to penetrate the growth medium, grip the root portion of the seedling,
and pull it out of the tray. The penetration and retraction of the fingers was
accomplished using air cylinders. This was successfully used for transplanting
cucumber (Cucumis sativus L.) and tomato seedlings under laboratory conditions.

Choi et al., 2002 developed a plant removal mechanism where a five bar
mechanism composed of a fixed slot, a driving link, a driven link, a connecting link,
and a slider was used for obtaining the same output motion to the tip of driven link, to
which pick - up pins were attached. A pin driver was provided near the pick - up pins
to operate the pins to pull out and drop the seedlings. It was reported that the feeding
rates of the pick - up pins varied from 30 seedlings min-1 to 60 to 80 seedlings min-1.

Luan et al., 2006 used a pneumatic mechanism, which picked up the seedlings
and continuously supplied the seedling trays. The feed rate of this type of pick - up
unit was reported to be 40 seedlings min-1.

Sena, 2006 developed a seedling pickup unit to adjust the seedling pick - up unit
based on size and configuration of trays.

Lawrence et al., 2007 employed a pneumatic cylinder to power the piercing

mechanism of the dibbling planter.

Kumar and Raheman, 2008 revealed that fully automatic vegetable transplanters
required either plugs or pot seedlings. As initial investment on production of plugs is
high and mechanisms employed for the removal of seedlings from the tray are
complex, use of pot seedlings seemed to be a better option.

2.4.2 Other plant metering mechanisms for automatic transplanters

Huang and Splinter, 1968 developed a gravity type automatic vegetable

transplanter for tobacco seedlings with grid cartridge carrying and feeding unit. The
cartridges were connected to a single feeding mechanism, which consisted of a rotor
29
and a notched rack. A continuous grid method intermittently moves the grid cartridge
in a transverse direction. The emptied grid cartridge is reset to the starting position for
reloading potted plants. The first cartridge at the bottom is connected to the feeding
mechanism for transplanting. As the first cartridge is emptied, it is dropped into the
cartridge carrier on the side of the transplanter. Then, the second grid is dropped to the
operating position. The operation is continued until all cartridges are emptied. The
grid cartridges with bottom plates were mounted one above other. As a grid matches
the drop hole in the bottom plate, the potted plants dropped from a certain height to
the ground. The parachuting effect of a plant leaves keeps the potted plant straight
upward during the fall. Suction force of 1½ lb can be applied at the drop hole to
accelerate the plant drop.

Boa, 1978 developed a transplant bandolier growing and handling system for
lettuce and celery seedling. Compressed peat modules were contained and connected
with plastic coated paper. The system included a module forming and planting
machine along with a tractor mounted automatic transplanting machine

Brewer, 1978 used a bandolier by joining two polyethylene strips by heat

welding to form cells in a continuous belt, in which grass seedling were grown and
automatically fed into a planter. The paper reinforced by plastic was used to confine
the growing medium. The grass seedling was removed from the bandolier belt by
separating the polyethylene strips just before planting.

Chow et al., 1980 designed a planting mechanism for a hand - fed lettuce
transplanter. It consisted of a planting head, two butterfly type, solenoid powered
gates, which allowed the planting of only one seedling at a time. During a planting
cycle, these gates alternated from the closed to the open position. This allowed a
ready positioned seedling to slide into the furrow, while the back ends of the gates
prevented the next seedling in the chute from sliding out. After the seedling was
planted, the gates closed and the next seedling slide into the ready position to
complete the cycle. The gate mechanism consisted of a pair of offset slider - crank
linkage that was attached to the gates and linked to a T bar. An electrical pull type
solenoid was used to activate T bar. For reciprocating T bar motion, a compression
spring was used with the solenoid. The solenoid was operated by a 17 Volts DC motor
to prevent heating. When the seedling hit the soil, its linear momentum changed to
30
rotational momentum, which caused the seedling to tip. The feeding chute of the
feeding machine was at an inclined angle of 30 .

Moden and Hauser, 1982 developed an automatically fed bandolier transplanter

for grass seedlings with horizontal feeding mechanism which carried the plant and
root plug in the horizontal position. The end of the bandolier belt was manually placed
onto a notched roller. A cushioned roller applied pressure to the top of the notched
roller to ensure positive contact between the belt and notched roller. The notched
roller was intermittently moved by a motor driven ratchet to feed the belt cutter. After
they were cut free of the belt, the plant cells dropped through an electronic sensing
device, which determined whether a plant was present in the cell. If no plant was
present, the plant cell dropped onto a reject gate and fell to the ground. The belt cutter
operated until a cell containing a plant was detected, and then stopped. The alignment
gate closed momentarily to align the plant cell in the horizontal mode, and then the
cell dropped onto the holding gate. As the plant cell dropped from the alignment gate
to the holding gate, the reject gate opened to allow it to pass. The holding gate
released the plant cell when a switch was closed by the feed trough cam leaver thus
filling the feed trough.

Moden and Hauser, 1982 developed an automatically fed bandolier transplanter

for grass seedlings with vertical feeding mechanism. The end of the bandolier belt fed
between two vertical, foam cushioned feed rollers. As the bandolier belt moved
between the feed rolls, the plant stems were grasped by two foam covered belts
located above the feed rollers. The plant cells and seedlings moved through the feed
rolls and foam belts until they are detected by an electronic sensor. The sensor was
located one plant cell width from the cutting blades. The bandolier belt passed
through the cutting mechanism as it leaves the feed rollers. When one cell passes
through the cutter, the sensor stops the feed or advance motor. This action started the
cutting motor which turned one - half revolution, causing the cutting mechanism to
close and cut the belt. The cutting blades were opened by a spring. The individual
seedlings were suspended from the foam belts by the stem. If no seedling was present,
the cells fell to the ground, thus no reject system was needed. This sensor halted both
the advance and cutting motors with one seedling in position to drop into the planting

31
dibble. When the dibbles reached the dead centre a micro - switch located on the
planter frame closed, thus starting the advance motor.

Lee et al., 1982 designed a seedling block transplanter with an intermittent

feeding mechanism. The feeding mechanism included a conveying belt, a Geneva
drive, a conveying chain, and a push - pull linkage. The seedling blocks were first
depotted and arranged in line on the carrying tray. The feeding belt moved the first
row of seedling block toward the transplanter head. When all the seedling blocks in
the first row have gone into the transplanter chute, the tray moved sideways so that
seedling in the second row was on top of the feeding belt. A pair of off - centre
pulleys was used to lift and lower the feeder belt. With the aid of auxiliary roller chain
connecting both pulleys, the belt is lifted vertically 50 mm higher than the slot in the
carrying tray. Thus the seedling blocks were raised above the tray and carried by the
belt by the time the two pulleys lower the belt, one row of seedling is carried away.
Then through the Geneva drive engagement, the carrying tray pushed by the
conveying chain to one side so that the next row of seedling blocks was moved into
place to begin the next feeding. To prevent slipping, the friction between seedling
blocks and the belt was increased and the friction between the tray seedling and the
seedling block at the exit was decreased. Coarse sand paper strips and a Teflon sheet
were attached on the belt and the tray exit respectively.

Shaw, 1986 revealed transplants are thought of as the small seedling plants that
are pulled from plant production beds and these were then commonly referred to as
bare root seedling. These bare root seedlings were usually lower in cost than the
modular transplants, but their survivability was usually not as good. The bare root
seedlings were being replaced by the more expensive but better quality modular
transplants. Modular transplants lend themselves to mechanised or automatic planting
where no manual handling of the plant is necessary except for loading trays of plants
into transplanting machine.

Burgoyne, 1987 experimented with plugs for their different characteristics as

shown in table 2.3.

32
Table 2.3 Characteristic of tomato plants grown in blackmore square cell and
round cell

No of plants Stem height Removal Root mass Stem mass

Cells/tray
tested (mm) force (N) (g) (g)

200 s* 15 182 1.39 8.0 3.5

288 s 25 155 1.37 4.0 1.5

288 r* 15 173 2.58 3.0 2.0

(*s - square and r* - round))

Munilla and Shaw, 1987 described a dibbling transplanter. It had six buckets
connected through the arms to a circular drum. The circular drum had specially
shaped cams that help in vertical descent (with zero horizontal motion) of the bucket
when the drum was rotated. The bucket received the seedling during its vertical
descent, made a hole in the soil, and opened to release the seedling in the hole. The
bucket closed again as it rose to accept the next seedling. A successful planting had
been defined as having seedlings inclined less than 30  from the vertical.

Rumsey, 1987 designed a conveyor type chute delivery system. The delivery
system was comprised of the chute frame, the chute conveyor assembly and the gate
mechanism. The chute frame was constructed from sheet steel with two sheets of
teflon covering the inside bottom of the chute. The teflon provides a low friction
surface for the movement of the seedlings. The frame was slotted to provide an
adjustable mounting for the gear belt pullers. The conveyor portion of the delivery
system ended 300 mm from the bottom of the chute.

Suggs et al., 1987 designed a self feeding transplanter with ferries wheel, where
one of a series of plant gripping wheel mounted hand pairs rotated into position,
closed on a plant and tore it from the strand along a line of between plant perforations.
After further rotation, the hands separated and released the plant into the open furrow.
A restraining device which opened to let the hands pass held the strands so that only
one plant was torn from the strand per cycle. The hands were opened and closed by
follower bearings rolling along a circular cam.

33
 Hergert, 1988 evaluated a Lennen planter and modified as Lennen RT2
transpanter for tomato transplants. The machines were designed for transplanting
plants grown in paper pots and RT2 was used for almost any type of tray plant. The
carousel or rotary cup bank consists of 10 cylindrical cups 60 mm diameter, angled at
the bottom, and closed by a hinged flap. As the carousel rotates, leavers on the flaps
opened each cup in turn, allowing the plant to drop through a delivery tube into the
furrow opener. The carousel speed was easily adjusted to provide various plant
spacing through changing a sprocket device in the drive from the packing wheels.

Shaw, 1988 developed a spatial kinematic mechanism that received a horizontal

seedling from the plant supply mechanism conveyor and turned it to an erect position
before inserting it into a dibbled hole in the soil. The individual plants or seedlings
were transported in cups from the loading position of the plant setting mechanism to
the point where they were inserted into the ground. The plants were discharged from
the cups through the pointed end of each cup, which also dibbled the planting hole in
the soil. The motion of each cup on the mechanism was such that the rearward
velocity of each cup relative to the machine matched the forward machine velocity so
that each plant was set in a stationary position in the soil.

Rumsey et al., 1989 designed a two row electro - hydraulic vegetable seedling
block transplanting. A specially designed seedling growing tray was an integral
component of the depotting operation. Each tray had a notched side bar which
interacted with indexing device in lining up the seedling pots with the extracting
assembly. When a tray was fed onto the tray conveyor, it moved horizontally until one
of the tray’s notches engaged the indexing device. Next, the extracting assembly
powered by a hydraulic cylinder pushed two rows of four seedlings half way out of
the tray. The seedling guide then advanced, providing a path for seedling to the chute.
After the guide was fully extended, the seedlings were pushed the rest of the way out
of the tray and the pushing shaft fed the seedlings onto the chute. Depotting
operations were repeated. Once the seedling entered the chute, they were conveyed to
the gate mechanism. They were released in 300 mm interval into a furrow made by
the plough. Once in the bed, the seedlings were buried by the furrow closer.

Lloyd Arnold, 1990 designed a tray placement system to support the trays
during the travel of the machine in the field, which took the tray from its storage area
34
and fed it to the seedling ejection mechanism and collected emptied trays. Tray
positioning was also done with an automated indexing mechanism to position the tray
for seedling extraction and ejection, missing seedling detection and compensation,
neighbouring seedling interference separation, root entanglement interference
separation, seedling extraction system utilizing grippers, seedling propulsion system
to furrow and furrow opening.

Huang and Ai, 1992 developed an air pruning plant tray system. Air pruning
totally eliminated root binding and controlled uniform shoot size by maintaining an
approximately constant shoot - root ratio of plants. The air pruning phenomenon also
resulted in a compact mass of roots which held together the media mass in the cell,
facilitating fully automated transplanting. The air pruning plant tray system included a
thin black plastic tray with truncated pyramid shaped cells which tapered upwards
with open bottom to provide root air pruning. The top edges of each tray cell were
rounded for easy passage of seedling without any damage to leaves during high speed
downward removal at the time of automatic transplanting. The tray systems are self
supporting and should be placed in growing areas above the ground to provide for air
circulation underneath to facilitate air pruning. This was accomplished by using
commercially available bench mesh or laying small diameter PVC pipes as two
supports crosswise under the tray systems. The air pruning tray system also provided
easy handling and shipping of fully grown seedling without damage because screen
bottom holds seedlings in the tray cells.

Nambu and Tanimura, 1992 developed a separating and transplanting

mechanism for a chain pot seedling. The chain pot was placed in position by hand and
the open planting fingers begun rotating downward. The stoppers were in the closed
position, the planting fingers closed around the pot, and simultaneously the stopper
opened. Then the planting fingers began to pull the chain of pots. The stopper closed
as soon as the planting fingers passed through. The next pot was stopped by the
stopper. As the planting fingers continued to rotate, the pot they were holding was
separated from the chain at the perforated paper joint. The planting fingers continued
to rotate and insert the individual potted seedlings into the field. Then the planting
fingers opened and the seedlings were planted.

35
 Osamu Sakaue, 1992 developed a robotic system for automating vegetable
seedling production and transplanting operation. It was inferred that the time required
for manual removal seedling is 1min 34 s per tray. Transplants with height of
170 to 186 mm, and diameter of 3.2 to 3.8 mm was used. The manual pulling force
was observed as 1.0 to 1.1 kgf and robot pulling force was 1.2 to 1.4 kgf for broccoli
seedlings with operating speed of 0.288 km h-1.

Suggs et al., 1992 designed the plant removal mechanism, where the vertical
rows of plants were pushed onto a spiked chain that carried them to the ground, where
they were spaced out in a furrow as the chain rolls around a bottom sprocket.

Marr, 1994 described types of transplanters called punch planters (automatic)

and water - wheel planters (semi automatic). The punch planters transplant through
plastic mulches by puncturing the mulch and the soil and setting the seedling into the
holes. The water - wheel planters were similar to the punch planters with the addition
of a large tank filled with water or fertilizer solution. As the hole was formed for the
transplant, a portion of the solution was injected into the transplant hole. Seedlings
were hand - set into the watered hole by operators riding low near the ground. As the
seedling was pushed in, mud from the bottom came up along the sides and covered
the top to complete the transplanting operation.

Kumar and Raheman, 2011 developed a feeding conveyor to carry the pot
seedlings in a rectangular array and feed a linear array of pot seedlings in upright
orientation on each slat into the space between the flights of the continuously
operating metering conveyor, without any damage and tilting of pots. The function of
the metering conveyor was to continuously push the pot seedlings in upright
orientation towards the seedling drop tube one by one and drop at the desired in - row
spacing. It was designed as a single stranded horizontal pusher type chain conveyor.
The feeding conveyor operated only for the time required to move one linear array of
pot seedlings on the slat towards the metering conveyor.

2.5 Furrow opener and closer mechanisms for vegetable transplanters

Huang and Splinter, 1968 used a spiked opener for dry field operation, and as
the opener was lowered to open a furrow, the centre and side spikes pulverized soil in
the furrow to form a soft - landing region for potted plants to absorb the impact and to

36
prevent tumbling. The press wheel was set at an angle of approximately 45 with
respect to the ground and slightly opened to the forward direction in order to obtain a
more normal closure of the furrow around the transplants without excessive
compaction of the soil.

Chow et al., 1980 devoloped a lettuce transplanter where the furrow opener
called Iseki hiller was fitted with extended side plates to prevent the furrow from
collapsing before the seedling was vertically positioned. The opener was adjusted for
a 152.4 mm wide furrow and with 203.2 mm high side plates. A 101.6 mm deep
furrow was made. The soil in the green house was very loose and dry, so simple
curved scraper plates were used to close the furrow. The furrow closers were attached
to the side plates of the furrow opener. They were adjusted to change the amount of
soil pushed into the furrow by changing their spacing and by varying their distance
from the ends of the side plates.

Moden and Hauser, 1982 used ferries wheel mechanism and four planting
dibbles were mounted on it. As the dibbles approached twelve o’ clock position, the
feed trough was raised from the horizontal to vertical position, thus feeding the plant
cell into the dibble. The rubber feed trough was pushed away by the rotating dibble
and the feed trough returned to the horizontal position. The holding gate was then
activated and another cycle started. A planting dibble which was similar to
mechanical dibble was developed. They entered and left the soil at zero velocity. The
plant and plastic encased root plug were dropped into the tubular top of the dibble.
The bottom of the dibble was closed by a hinged trapdoor which remained closed
until the dibble had penetrated to the planting depth, at which time it opened and left
the plant behind in the hole. Packing wheels followed behind the dibble units to
compress the soil around the root plug.

Hergert, 1988 developed a planting system for Lennen RT2 transpanter. The
planting system consists of a stainless steel furrow opener and a unique plant guide
consisting of long spring tines protruding from two V- belts driven from the packing
wheels. The seedlings dropped from the delivery tube into an area near the back of the
furrow opener, bounded by a metal tongue at the front and the tines at each side. The
tines kept the plant upright and allowed soil to filter through to cover the plant root

37
ball. The tines gradually withdrew from the ground just before the packing wheels
firm the soil around the plant

Pelletier and Hergert, 1988 evaluated a Regero HD B4T transplanter with

tomato transplants. As the transplant fell into the cup, the root ball sits between each
side of the furrow opener, when the mechanism was activated. A scraper on the front
of the cup pushed the root ball out of the shoe and cleaned all soil from the furrow
opener. Shorter plants required the action of the retracting door to keep them from
falling over during the planting process. When the door was removed, the plant relied
on the foliage to hold it upright in the cup.

Lloyd Arnold, 1990 concluded that the delivery tube in relationship to the
packing wheels and forward ground speed were found to be important factors. With
faster planting speeds, the delivery tube was moved away from the packing wheels
towards the front of the shoe because soil was flowing around the heel of the shoe
faster and this adjustment allowed the plug to reach the bottom of furrow. Slower
ground speed had the delivery shoot being moved back toward the packing wheels
away from front of shoe so that the plug had less time to tip over in plant furrow as
soil was not flowing in around the plug as fast.

Suggs et al., 1992 concluded that the space between the two press wheels was to
be reduced to 30 mm and an additional weight of 20 kg was to be added to the rear of
transplanter. Capture and retention of the plant on the planting rod was improved by
replacing the rod tip with three needles 60 mm long , 1.5 mm in diameter spaced on a
triangular base so the wires were parallel and about 50 mm apart. Thus, there were
three points per planting rod instead of one onto which the plant roots could be
captured. The tip speed of the planting rods around the bottom end of the guide was
approximately equal and opposite in direction to ground speed so the plant went in to
the furrow with essentially zero velocity.

Shaw, 1997 designed an automatic transplanter where the in - row spacing of

the seedling plants could be adjusted by changing the distance between and the
number of cups on the plant setting mechanism. The planting depth was adjusted by
use of the hydraulic lift on the machine wheels. Two pneumatic press wheels were
used to firm the soil around the seedling after they were set into the soil.

38
 Chaudhuri et al., 2002 used a runner type furrow opener and two covering
wheels to close furrow and compact soil around the plant. Depth adjustment was
made with the help of two pneumatic depth wheels.

Srivastava et al., 2006 used a pair of press wheels tilted outwards at the top is
the most commonly used device for closing the furrow and compacting of soil.

Satpathy and Garg, 2008 used runner type furrow opener with soil compacting
unit tilted at 15 .

Kumar and Raheman, 2010 reported that various types of furrow openers were
used in transplanters. Disc coulters are used to cut the trash before opening the
furrow. In general, a double shoe (straight runner) type furrow opener is the most
commonly used furrow opener in transplanters for use on well prepared dry and loose
soil. However, in transplanters for use on raised beds and nurseries where soil is
moist, a shovel type furrow opener could be used. Soil covering devices are used in
transplanters provided with shovel type furrow openers. Double disc type soil
covering devices are used with such transplanters.

Kumar and Raheman, 2011 used a PVC pipe as the seedling drop tube and it
was provided at the seedling discharge point of the metering conveyor. A quick return
valve was provided at the mouth of the seedling drop tube just below the base plate of
the metering conveyor to ensure that the pot seedling falls in upright orientation
through the tube into the furrow. It was a sliding type valve and operated by a lever.
A reversible shovel type furrow opener with front face width of 450 mm was selected.
A pair of blades converging at the rear end was used as the soil covering device. The
blades were provided on each side of the seedling drop tube.

Narang et al., 2011 developed a semi automatic transplanter where the furrow
opener had a length and height of 535 and 230 mm respectively. The furrow width
made by furrow opener was 100 mm. The packing wheel made of mild steel of 3 mm
thickness. The sheet was bent to a diameter of 520 mm from one side and 555 mm
from the other side. The distance between the two wheels at the bottom end was
95 mm and 175 mm at the top end, when tilted at an angle of 75  with the horizontal.
Hardened plug seedlings were loosened from the trays before transplanting

39
by the operator. The depth of planting was varied between 50 and 56 mm and the
plant to plant spacing was 400 mm.

2.6 Seedling and plug characteristics

Huang and Splinter, 1968 developed a tobacco transplanter that automatically

fed 50.8 mm square by 50.8 mm deep soil blacks containing seedlings. He compared
growth rate in the field and in the growth chamber in order to compare conventional
transplanting and potted plants. The result indicated that the potted plant transplanting
resulted in a higher leaf area as compared with the conventional transplanting. This
was due to the fact that the root system of the potted plants was undisturbed at the
time of transplanting.

Chow et al., 1980 designed a hand- fed lettuce transplanter (Latuca sativa L.)
seedlings that had 82.6 mm diameter and either 50.8 or 76.2 mm tall soil blocks. They
used 6 week old seedlings grown in soil blocks made of 2/3 mica - peat and 1/3 manu
silt loam soil.

Penley, 1981 used a tray made of a fibrous material which also served as the soil
medium. These trays were separated into seedling containing blocks that were
planted.

Lee et al., 1982 designed a seedling block transplanter for lettuce. Seedlings
were grown in 51 mm tall, 51 mm inside diameter, bottomless, reusable PVC pots. Six
week old seedlings leaves were fragile and had a projected area of approximately 120 mm
in diameter. Plant height was about 140 mm measured from the bottom of the block.

Hergert et a1., 1988 developed a rack system for transporting tomato seedling
from green house to field. The racks, either a 2240 x 1120 mm wooden racks with
wires stretched across the wood member or steel 1 x 1/8 th angle iron constructed held
16 trays. The common tray used for tomato transplants was the blackmore 288
(or equivalent) tray which held 288 plants in individual cells. The seeded trays were
placed on the racks and were handled that way throughout the growing process.

Brewer, 1994 concluded that the seedlings which are manipulated by machines
should be short and sturdy. No growth regulator can be used on vegetable plants and
brushing was used for controlling height. He used tomato or pepper seedlings about
150 mm tall in this study. Trays contained 200 cells spaced at 38 mm on center in a
40
10 x 20 array. Modules had the same basic dimension as the trays, but are 250 mm
deep.

Singh et al., 2007 optimized cell size and shape for raising tomato seedling and
found that growth rate always tended to be proportional to the volume of the container
cell. Soilless media for raising seedling in plug trays included coco - peat, vermiculate
and perlite (3:1:1 v/v) was used. The root proliferation was clear in the inverted
pyramid shaped cell grown seedlings as compared to the round shaped cell grown
seedling. The root development in the round celled seedlings was mainly located in
the center of the root ball and some lateral roots grew to the outside of the root ball.
The stem diameter was 1.9 to 2.9 mm for inverted pyramid shaped cells and 2.1 to
3.1mm for round shaped cells. The true leaves were found to be 2 to 3 numbers.

Singh et al., 2010 evolved a method of nursery raising technology in which off
season seedlings were raised in small polyethylene bags using soil and compost 50:50
as a media. It was reported that this system needed a lot of protected space and was
labour intensive. During the transportation process, the crop in main field needed
large quantity of media and had high damage to root system, therefore not accepted by
growers. A nursery raising technology was developed where cucurbits were raised in
multi celled plastic plug trays containing each cell volume of 18 - 20 cc and inverted
pyramid in shape by using coco - peat, vermiculate and perlite as soilless media in
3:1:1 ratio on volume basis. This technology was efficient in vigorous root
development, but also suitable to avoid any damage to the roots and shoots of the
seedling at the time of transplanting. The seedlings of these cucurbits were mostly
ready for transplanting within 18 - 22 days after sowing.

Kumar and Raheman, 2011 used paper pots seedlings for mechanical
transplanter. The seedlings were grown in aluminium trays in a double layer
polyethylene - covered unheated greenhouse. The ready - to - plant paper pot seedling
of tomato was 21 days old after seeding and was given 4 day of hardening. Each
seedling was grown in a double layered cubical paper pot of 500 mm3 volume filled
with vermicompost, soil and sand in the proportion of 1:1½ :1½ by volume.

41
2.7 Performance of the vegetable transplanters

Huang and Splinter, 1968 designed a three point hitch tractor drawn automatic
tobacco transplanter which transplanted four plants at a time, that is, two plants on
each row spaced 560 mm, apart with a distance of 1120 mm between two rows. It was
operated at a maximum speed of 4.8 km h-1 and to be driven by the power taken from
the tractor wheel. The planter carried four grid cartridges of each 220 plants totalling
880 plants. As the tractor moved forward 1120 mm, the feeding mechanism moved
the cartridges 50 mm longitudinally, so the four loaded grids match the four holes of
the bottom plate. The four plants fell freely to the ground and thereby transplanted.

Todd, 1972 found that the quality of seedling increased over bare root type and
made handling of seedling easier, 200 to 300 plants can be handled in each tray.

Chow et al., 1980 evaluated a hand fed lettuce transplanter in the laboratory and
field with the following variables speed, angle, dwell, compression spring force and a
furrow closer adjustment. The plant spacing was 384.8 mm. The transplanter was
operated at a speed of 0.64 km h-1, 0.524 rad chute angle and 0.62 dwell, 300 mm
spacing found less than three per cent planting error with the planting rate up to
9500 plants h-1. The seedlings tend to tip over when its sliding speed was much
greater than the transplanter speed. To prevent tipping, the edge of the transplanter
head was bevelled, a smooth furrow bottom, the minimum pivot angle was
subjectively set at 0.471 rad. The tipping was inhibited by having two seedlings which
stay in contact for 25.4 mm of seedling travel before the gates stopped the second
seedling. An improved gate design prevented tipping of the 76 mm tall seedlings
when the chute angle was less than 0.698 rad. At steeper chute angles, tipping was
prevented with the use of rubber finger. The solenoid force was insufficient to keep
the gates closed when the seedling impacted on them. Round bumps were placed at
the front of the gates to prevent this type of misplaced planting.

Moden and Hauser, 1982 used transplants that were grown in hand made plastic
bandolier cells of cylindrical shape. The plastic encased plant cells (root plugs) were
60 mm long and 8 mm in diameter (small) and 32 mm diameter by 114 mm long
(large). The bandoliers were made from 0.051 mm thick polyethylene plastic. The
bandolier cells were filled with a commercial potting mix containing peat and
vermiculate. The plant spacing was 280 mm.
42
 Lee et al., 1982 conducted field test of the transplanter in an open ended
greenhouse with a hydraulically operated prime mover. The feeding rate was set at
15 seedlings blocks per 20 s. It was operated at 0.8 km h-1 and the plant spacing was
304.8 mm. Seedling blocks were pushed easily into the transplanter chute as long as
the soil blocks maintained their original shape without deformation.

Branch, 1985 developed fully automatic transplanters for lettuce and celery
transplants. The capacity of an 8 row transplanting machine was reported to be
33, 000 plants h-1.

Shaw, 1986 reported that the tray was the best plant handling system because
they are convenient packages of plant and the plant density that only 12 to 15 trays
were needed to plant the typical row of tomatoes in field. Also trays can be inspected
for missing plants and replacements made in the green house. Replacing blanks in the
trays were carried on in the green house over an extended period of time and vary
possibly automatically with a robot.

Suggs et al., 1987 evaluated automatic feeding ferries wheel transplanter. The
transplanter was carried and propelled by a 2 wheel garden tractor. The lettuce
seedlings were planted in the centre of raised beds about 200 mm high and 510 mm
wide at the spacing of 600 mm and operated at the speeds up to 100 plants min-1.
Properly hardened lettuce plants about 38 mm tall were planted without any damage
while planting with 50 to 75 mm tall were severely damaged in the restraining device.
The taller plants were not rigid enough to remain in a horizontal position during the
feeding operation. The same result was found in the sweet corn plants over 100 mm
tall. Sweet corn plants 64 to 75 mm tall were easily planted. The automatic roll
feeding transplanter evaluated implied that the machine was operated at 84 to 140
plants min-1 at the spacing of 610 mm. The automatic feeding bare root transplater
could be operated at nominal speeds of 15 and 43 plants min-1 with the plant spacing
of 610 mm.

Feldman et al., 1988 evaluated the castle and cooke transplanter, in the field
which was a well prepared, smooth, bare, sandy bed as used for transplanting tree
seedling. Soil was moist, but not sticking to soil working parts of the machine. The
tray used was the blackmore 288 with either round or square cells. The plants were
120 to 200 mm tall when transplanted. It was operated at the speed of 2 to 6 km h-1.
43
About 9 to 10 trays were run through the mechanism of one row of the 8 row
machine. The transplants used were approximately 50 to 70 mm tall on top of a 10
mm diameter by 45 mm high plug. Plant spacing was about 76 mm and row spacing
of 150 mm (as required for trees), but double that spacing when planting from the
trays with plants in every second cell. The planting rate achieved was about
2, 00,000 plants h-1. While planting with tomato seedling spaced at 350 mm, at the
same forward speed as for tree seedlings, the planting rate for tomatoes was 71 plants
min-1 row -1
. It was observed that the plants should be taller. Still, there were
indications of leaf interference problem in the machine. Empty tray containers were
removed from the machine and full ones installed, using a fork lift. The operation is
part of the machine turning cycle at one end of the field. The planting rates of cup
machines with tray plants achieved a rate of nearly 70 plants min-1 row-1 compared to
65 plants min-1 row-1 at most, using gripper machines with bare root plants. This
since was cup machines needed only one worker per row, instead of two as on the
gripper machines. The machine planting rate reduced slightly compared to bare root
seedling when using tray plants in gripper machines. Conventional machines using
bare root transplants and 2 operators per row was about 61 plants min-1. Comparing the
M - 4000 with racks to conventional bare root operations, the time to plant 1000 plants
-1
row was reduced by more than 7 min. The force to remove the plants from the
round cells was 1.8 times that for the square cells.

Hergert, 1988 evaluated the performance of a Lennen RT2 transplanter model.

The working parts of the planter were mounted through a floating parallel bar linkage
with a clamp on a tool bar. Each planting unit had independent depth adjustment
controlled by a hand wheel. Each planter unit was about 500 mm wide. The unit was
capable of planting up to 120 mm deep, with 180 mm tall tomato plants. The
observations summarized that through plug plants are more expensive to purchase
than bare root plants, there are several operational advantages with plug plants
especially in the areas of labour saving and plant survivability.

Pelletier and Hergert, 1988 evaluated a Regero HD B4T transplanter with

tomato transplants. The machine was tested in a field having permanent beds formed
at 1700 mm on corner. Planting depth was adjusted to approximately 75 mm with the
packing wheels set as close as possible to the cup mechanism. The tomato plants were

44
approximately 150 to 200 mm total length, including a root ball approximately 38 mm
long. The plant spacing was 400 mm. The planting rate, for one person feeding two
rows was limited to 30 to 35 plants min-1 row -1. The packing wheels were adjusted as
close as possible to the furrow opener.

Rumsey et al ., 1989 evaluated a two row electro hydraulic vegetable seedling

block transplanting. The hydraulic pumps were driven by the tractor’s PTO shaft and
the electrical system was powered by 12 V battery. Results of these tests inferred that
a tray of 32 seedlings could be depotted in about 40 s enabling the trasplanter to have
a forward speed of almost 0.5 km h-1 with a plant spacing set at 302 mm. Out of 224
seedlings tested, two received leaf damage and one received soil block damage. Soil
blocks used in this experiment were made from 60 per cent mica peat and 40 per cent
manu silt soil by volume. It has a field efficiency of 75 per cent and that the spacing is
950 mm, the effective field capacity of the transplanter would be 0.05 ha h-1.

Lloyd Arnold, 1990 concluded that the per cent tray fill (number of healthy plug
plants per tray) had a direct effect on planter performance as it did not have the
capability of detecting and compensating for missing plants in the tray cells .
A missing cell (cel1 containing no plant) or small undersized plant resulted in a gap in
field stand because of its absence or the small seeding was buried in the plant furrow
by the packing wheels. It was observed that works on missing cell detection and
compensation, the optimum tray design, cell size and plant density for plug transplant
production was still not known. The reduced weight of the root ball associated with
these smaller seedlings may present problems of delivery to planting furrow by
gravity. The seedlings growing uniform quality plug plants are mandatory for
transplant automation to succeed.

Nambu and Tanimura, 1992 evaluated an automatic transplanter for lettuce and
cabbage seedling at the operating speed of 0.58 m s-1 and the intra row spacing was
300 mm and row width 600 mm. The machine worked at speeds upto 120 plant min-1.

Suggs et al., 1992 concluded that there was some problem with the transfer of
plants from the tray to the planting rod. This was due to the fact that the available
plants did not have well developed root balls. In many cases, the soil media mass
crumbled when the plant was pushed from the tray making it difficult to retain the
plant on the planting rod. The indexing mechanism, which moved the tray laterally to
45
bring the next row of the plants into alignment with the push rods sometimes
malfunctioned due to the slight non-uniformity in tray design and shock of the tray
being rapidly decelerated at the end of the return stroke. Plant spacing varied
significantly because some plants dropped off the planting rod prematurely. The
mechanism operated effectively at speeds upto 3 cycles s-1. The plant spacing was
677 mm with a standard deviation of 26 mm.

Marr, 1994 opined that the transplanter has to be operated at a speed that allows
careful placement and attention to problems that develop.

Brewer, 1995 found that a spindly, limp top could allow a seedling to drop in
such a manner that the whole seedling would rotate and be planted upside down. An
adequate top with some strength act as a sort of rudder to keep the seedling oriented
vertically throughout its fall. When the top was too stiff causing the seedling to hang
up in either the shipping module or one of the cassettes and the top was full, causing it
to be caught by the bottom flaps on the shipping module. When a square shaped
seedling plug was turned 45 deg with the square shaped cell in the shipping module,
the seedling could hang up and not fall to the cassette below. An occasional seedling
would enter the dibble, then will not exit when the dibble opened, but would ride
around again and be planted with the next seedling entering that dibble.

Brewer, 1997 found the hand feeding of the Model - 5000 transplanter was
limited to well under 2 seedlings s-1, whereas automatic feeding can reach at least
5 seedlings s-1. It was inferred that humans unable to feed seedlings much faster than
one seedlings s-1 without making errors, which cause skips in the field. Once the error
started to occur as a result of operators not being able to respond quickly enough to
the rate of transplanting. Errors increase exponentially as the rate of transplanting is
increased.

Shaw, 1997 designed a transplanting machine for both tomato and pepper
seedlings into bare ground and through plastic mulch. The transplanting rate was
7000 plants h-1 row -1
. The machine could be adopted to other trays that had a
rectangular cell pattern with minimum drain hole diameter of 56 mm. The machine
was designed for 180 to 200 mm total length seedling and some difficulty was
experienced with plants over 230 to 250 mm in length. If plants developed past the
ideal transplanting stage, there was root growth between the cells and into the
46
polystyrene tray material. This transplanter can empty a typical tray every two and a
half minutes.

Shaw, 1999 developed sequential plant ejection system to eliminate the

handling of seedling enmesh that contributed to the entanglement problem. The
sequential plant removal system, individual plant ejection plungers were aligned with
the drain holes in a row of tray cells and actuated in rapid sequence starting at one
end. The entanglement between seedling was reduced because the plants below and
on one side of the one being ejected had already been removed from the tray and the
moving foliage lifter was lifting of the plants above.

Tsuga, 2000 evaluated the fully automatic transplanter by using the cell mold
and pulp mold pot cabbage seedling. These machines fed vegetable seedling
automatically and enabled continuous transplanting work on 2 ridges simultaneously
at a planting speed of 60 hills row -1 min -1. The working speed was 0.33 to 0.39 m s-1
at the hill spacing of 300 and 600 mm row spacing with one operator.

Labowsky, 2001 suggested that operators should not be so involved in placing

plants in the machine that they cannot watch for problems that develop. A rotary cup
type planting unit on a semi automatic transplanter allowed for higher forward speed
than that of a pocket - type planting unit.

Chaudhuri et al., 2002 evaluated an imported transplanter used for transplanting

tomato and chilly seedlings. The average seedling height during transplanting ranged
from 160 - 250 mm. The transplanter was operated at the forward speed of 0.23 to
0.33 m s-1 with the 300 mm plant to plant spacing and 600 mm row to row spacing. It
was observed that for proper covering of seedlings their length during transplanting
should be greater than 150 mm but not more than 200 mm. The covering was not
satisfactory under moist condition. The machine was suitable for use with bare root
nursery seedlings.

Anonymous, 2004 reported forward speed as 0.9 km h-1 and field capacity as
0.1 ha h-1 for planting tomato at a 600 mm row spacing and 450 mm in - row plant
spacing using a tractor drawn two row semiautomatic transplanter with pocket - type
planting unit.

47
 Anonymous, 2004 found that in semi automatic transplanter each planting unit
generally requires two persons for placing seedlings in pockets. Reduction in speed of
operation and high per cent of missing seedlings was observed when operated with
one person per planting unit. The field performance of semi automatic transplanter is
presented in table 2.4.

Table 2.4 Field performance of two row semi automatic transplanter

Parameter Vegetable

Chile
Tomato Cabbage Cauliflower
Pepper

Row spacing, mm 670 670 670 670

Plant spacing, mm 250 - 300 500 - 540 250 - 300 250 - 300

Speed of operation, km h-1 0.90 1.10 0.95 0.90

Field capacity, ha h-1 0.082 0.090 0.092 0.084

Plant mortality, per cent 5.0 - 7.0 5.0 - 7.0 1.0 7.0

Labor requirement per row One person

Missed plantings, per cent 3.0 - 8.0

Anonymous, 2004 reported an average forward speed of 1.4 km h-1 and field
capacity of 0.14 ha h-1 for planting tomato, cauliflower, chile peppers, and eggplant
using a three row semi automatic transplanter with rotary cup type plantings.

Anonymous, 2006 reported that the capacity of semi automatic vegetable

transplanters for bare root seedlings varied from 35 to 45 seedlings min-1.

Satpathy and Garg, 2008 evaluated the performance of a two row semi
automatic vegetable transplanter at different speeds, soil moisture and seedling age
with respect to plants missing, planting angle, planting depth, plants in lying down
position and plant mortality for two vegetable crops viz. Tomato and chilli. A semi
automatic vegetable transplanter with the seedling age of 5 weeks for tomato and
17 - 19 weeks for chilli crop was evaluated. Plant missing was within the acceptable
limit of 3 - 4 per cent at an average speed of 1.0 to 1.2 km h-1 with two operators
feeding seedlings row-1. With two operators row -1
the plant missing reduced
48
considerably as compared to one operator per row. The transplanter was operated at a
spacing of 300 x 300 mm plant to plant and 670 x 670 mm row to row. It was
observed that planting depth increased with increase in soil moisture content and
seedling age. The average depth of planting varied from 23.3 to 53.2 mm in tomato
crop and 23.1 to 51.6 mm in chilli crop. The planting depth at 6 per cent soil moisture
content and seedling age of 4 weeks and 15 weeks for tomato and chilli crops were
found very low. The soil compaction was satisfactory at 10 per cent soil moisture
level and the lying down plants and plant mortality was the lowest. A length of
150 - 200 mm seedlings was used for this study. Plants angled at less than 30 with
the horizontal plane were considered lying down. The plant uprightness was found to
be 93 per cent.

Kumar and Raheman, 2008 tested the paper pot seedlings for use in a fully
automated array type vegetable transplanter and it was found that they can be easily
fed to, and metered in, the conveyor type units of the transplanter.

Manes et al., 2010 developed a semi automatic transplanter for vegetable

seedlings and evaluated in the field. Two persons, one for each row was required to
place the seedling in the rubber flappers when these open at the top position. The root
side of the seedlings was kept towards the operator. He used bare - root type nursery.
All the vegetable seedlings were transplanted using the machine at row to row spacing of
670 mm instead of 600 mm since it was not possible in this mechanical transplanting due
to high tread width of available tractors. The machine transplanted different types of
seedlings precisely and without any problem at a speed of about 1.0 km h-1. It was
observed that a person could put 30 seedlings min-1 in the picking fingers.

Kumar and Raheman, 2011 found that the labour requirement for transplanting
tomato seedling using the automatic vegetable transplanter was 103 man - h ha-1. This
included the labour requirement of 25 man - h ha-1 for refilling of missed plantings
and correct placement of tilted plantings.

2.7.1 Effect of field moisture content

Lloyd Arnold, 1990 found that watering the tray prior to planting would provide
a heavier plug to be extracted from the tray result in a faster drop time to the planting
furrow and shoe by gravity. Soggy over watered plugs with insufficient root

49
development allowed for poor extraction as the plunger pushed easily through these
root balls. If the soil was loose and friable, the soil flowed in too fast around the root
ball before the plant properly oriented itself in the furrow. The rubber fingers that
wiped off the needles were adjustable to account for variations in length of plant stem
and foliar growth. Root entanglement with root hairs growing between the tray cells
impeded planter performance. The tray filling process ensured that soilless mixture is
not over applied to the tray. Root development and firmness of the root ball was
important. Smaller plants which had not developed an extensive root system in the
tray cell, where the plunger pushed through the ball resulting in poor extraction and
delivery to planting furrow.

Satpathy and Garg, 2008 reported that, the field was irrigated and prepared at
16 - 18 per cent moisture content and then periodical checking of soil samples was
done to attain the desired level of soil moisture. The moisture content of 10 per cent
was found to be better with respect to plant missing, laying down plants, upright
plants and plant mortality. Lower per cent of upright plant at higher moisture content
was due to lesser soil flow towards the planting furrow, which resulted in poor soil
coverage. On the other hand, at lower moisture content of 6 per cent there was closing
of planting the furrow before the plants were released due to excessive soil flow.

Kumar and Raheman, 2011 evaluated a 2-row hand tractor operated automatic
vegetable transplanter for transplanting paper pot seedlings of tomato at 450 x 450 mm
spacing. The soil moisture content was 9.0 ± 2 per cent (dry basis). Twenty - five days
old pot seedlings of average height 120 mm and potting mix moisture of around 5 per
cent (dry basis) were selected. The hand tractor was operated at an average forward
speed of 0.9 km h-1. Fifty - four pot seedlings were placed on each feeding conveyor
(48 seedlings in the form of 6 x 8 rectangular array on the feeding conveyor and
6 seedlings on the metering conveyor up to seedling discharge point). Missed
plantings were filled by manual labour. Thus the tray carried a total of 108 seedlings.
The operational parameters, moisture content of the potting mix, plant height, linear
speed of metering conveyor and the ratio of linear speeds of feeding and metering
conveyors affected the working of the major components of automatic vegetable
transplanter.

50
Materials and Methods
 CHAPTER III

MATERIALS AND METHODS

In this chapter, the conceptualization, experimentation and development of the

final prototype automatic vegetable transplanter is explained. The experimental
procedures adopted to optimize the variables that would influence performance of the
contemplated automatic vegetable transplanter were conceived primarily to eject the
seedling from the protray, transport to the ground, plant it upright in the cleaved
furrow and compact it around without any damage to the root ball. The experiments
were planned for optimizing parameters influencing as plant removal, plant
straightening height while falling by gravity, root ball weight loss while falling to
ground are described. The optimization of operating speed, planting depth,
compaction, wheel angle and moisture content of the field are also described. The
development of tractor operated prototype automatic vegetable transplanter and its
performance evaluation in the field are also explained in detail.

The process of experimental development of the prototype vegetable

transplanter is discussed under the following subtitles.

3.1. Conceptualization of the machine process of transplanter

3.2. Experiments on seedling ejection from protray by gravity

3.3. Experiment on seedling ejection by clawing out from the top of the protray cell

3.4. Experiments on popping out the seedling from the back of the tray

3.5. Optimization of agronomical parameters of the growth media

3.6. Pulling force measurement

3.7. Drop test for weight loss of root ball and straightening height

3.8. Development of a prototype vegetable transplanter

3.9. Field level optimization of variables for the prototype vegetable transplanter

3.10. Development of the automatic transplanter mechanism

3. 11. Development of the tractor mounted automatic vegetable transplanter

3.12. Field evaluation of automatic vegetable transplanter

51
3.1 Conceptualization of the machine process of transplanter

The bare - root seedlings were deemed not suitable for automatic vegetable
transplanter, since of the intricacies in picking/placing them. So the vegetable
seedling grown in a protray was deemed easier for transplanting with machines. In
developing an automatic vegetable transplanter, two important factors were
considered. The first factor is the seedling removal which needs to be automated,
where seedlings need be removed from the protray one by one without damage to the
seedling stem or root balls. The mechanism of ejection is intricate and two methods
were contemplated, one where the plug is released by weight of the root ball and the
other where it is ejected by force. The concept of feeding the protray itself would vary
in these methods. The next factor is the transplanting of the ejected seedling to the
ground. Some earlier designs have used conveyor exclusively to convey the seedling
down to the ground. This process is very complex and mechanism intensive. Hence it
was decided to drop the released seedling to a check valve at ground level. The
seedling while dropped from the protray through a certain height needs to straighten
up when falling to the ground. But gravity converts the potential energy of the plug
plant into kinetic energy and when the plug plants strikes the ground or the valve, the
impact may cause damage to the plug, which was later planned to be cushioned by
some means. The straightening and impact force can be controlled by the height of
drop and the seedling could be planted without damage.

A narrow furrow was to be cut by a narrow cleaving opener and the seedling
placed upright in the cut furrow at the desired spacing on - the - row. After placing the
plug seedling, the furrow was to be closed and compacted by appropriately placed
compaction wheels, yawed and rolled to provide the closing/compacting action.

3.2 Experiments on seedling ejection from protray by gravity

Here the protray seedling instead of being drawn from the top was planned to
drop through the bottom by gravity. Open bottom protrays are not available
commercially. So protray cell bottoms were manually cut and used on trial basis. Instead
of using the whole protrays, protray was segmented into seven linear strips and used, so
that each strip with seedling can be fed to the machine rather than the whole tray.

The protray strips were cut manually to uniform size and the bottom was
removed manually for which a die was made and placed on top of the protray and

52
knife was used to cut the bottom of the protray. Figure 3.1, plates 3.1 and 3.1. a shows
the details of the open bottom type protray segments. It was felt that if this method
proved successful, the seven segments could be made attachable to each other to grow
the seedling in a whole unit, after which they could be separated and fed into the
designed machine. The important difference here is that the protray segment was
placed upside down, media filled and seeds sown in its bottom-up position. When the
seedlings are fully grown, the taper available in the cell will allow the seedling to
slide down easily by gravity.

3.2.1 Experimental test rig for testing the gravity ejection

The machine was to have a feeding chute which would allow the segmented
protrays into the machine from waist height. An experimental feeding chute was
developed to test the process by feeding the protray strips towards testing whether
gravity ejection is really feasible. The details of the chute are shown in figure 3.3 and
plate 3.2. Since the vegetable seedlings were grown in the protray strips, a transfer
plate was also developed for transferring the protray segments with seedlings into the
feeding chute (Fig.3.2. and plate 3.1.b). The protray segment with seedlings was
transferred to transfer plate. While transferring the seedlings to the transfer plate the
seedlings are loosened slightly by gravity from the protray due to the two round rods
provided at the inner edge of the transfer plate and the clearance created thus. The
transfer plate was then inserted onto the top of the feeding chute, the seedling tray
pushed down by hand, so that it slides downward until it reaches the bottom of the
feeding chute. At the bottom end of the parabolic sliding feeding chute, a matching
hole was provided. When the protray cell bottoms are synchronized in position with
the bottom hole of the feeding chute, seedling could fall down by gravity to the
ground. To prevent tipping of the seedling to side while moving down the chute, ribs
were provided on both sides of the feeding chute with a height of 15 mm.

3.2.2 Mechanisms for moving the protray cells in alignment to the hole

A 12 V DC, 50 rpm centre shaft gear motor was used for providing the
movement to the seedling segment. A hub with six equally spaced fingers was fitted
on the motor shaft. When the motor shaft is rotated, the fingers on the wheel moves a
linear distance of 36 mm (i.e. the centre to centre distance of the tray cells) and the
protray with seedlings was pushed forward down the rig such that each cell aligns

53
with the hole sequentially. The fingers were sheathed with rubber sleeves to avoid
damage to tray wall while pushing it forward. The details are shown in figure 3.3.a
and plate 3.3. The circuit that allowed the finger wheel to push the tray cells one by
one is shown in figure 3.4. A limit switch was provided which stops the movement
when each of the fingers moves the tray segment through one cell, hits the limit
switch and stops the movement. When the hole is aligned to the seedling bottom, it is
free to fall through it. A push button was connected in parallel to override the limit
switch when next movement was required. Thus the tray segment carrying the
seedlings was pushed one by one to the hole for making them to fall through.

The motor is connected through the NC (Normally closed) terminals of the limit
switch to the battery. So when the finger activates the limit switch, the circuit is
opened and the motor is stopped and is retained in that position. But when the push
button is pressed manually and momentarily, the circuit is completed and the motor
rotates till the next finger hits the limit switch and so on.

The mechanism was not smoothly moving the protray segment, since the fingers
were pushing the cells tangentially. Thus while moving the cells axially along the test
rig, they pierced into the plastic wall of the cells causing damage. Hence it was
decided to replace the mechanism with a better one.

3.2.3 Walking beam mechanism using double crank with belt drive

This is basically a double crank mechanism with the driven and driving radius
being equal. The geared motor drives one of the crank which is a toothed pulley of 1mm
pitch. On the top of the pulley, a crank pin was setup which connects to the driven crank
through a connecting rod of flat mild steel bar. The driven crank is also a toothed pulley
of 1 mm pitch and carries a crank pin on its top to connect the connecting rod. The two
pulleys are meshed with a 1 mm pitched toothed belt of 8 mm width and 200 mm length.
When the motor rotates, the crank mounted on its shaft drives the driven crank pulley at
the same angular displacement. The connecting rod hence moves in synchronization with
the pulleys and the connecting rod always moves parallel to the line connecting the
centres of the double crank. A push rod mounted perpendicular to the connecting rod
moves with the connecting rod with its end making a circular motion. The crank radius
being 18 mm, the push rod moves in a circle of 36 mm diameter.

54
 14 nos
A

44
36 A
530
24.5

32

Isometric view
0.20

SECTION A-A
SCALE 1 : 1

th
Fig 3.1 Open bottom pro - tray segment (1/7 of a full tray)

6
15

540
45

All dimensions are in mm

Isometric view

Fig 3.2 Transfer plate

55
Plate 3.1 Open bottom type protray (7 segments make one full tray)

Plate 3.1.a Open bottom protray segment

Plate 3.1.b Transfer plate

56
Waist level of a human feeding the protray segments
45

°
56

2240

1157
F

15
Isometric view

257
1370

Finger - wheel

DETAIL F Geared motor

Fig. 3.3(a)
All dimensions are in mm
Fig 3.3 Test rig for seedling ejection by gravity

57
Fig 3.4 Circuit diagram for imparting movement to the protray segment

58
Plate 3.2 Experimental test rig for seedling ejection by gravity

Plate 3.3 Finger wheel moving mechanism

59
 The push rod moves into the gap between the protray cells at one top dead
centre of the mechanism, moves the cell through 36 mm axially along the test rig
pathway and releases out of the gap at the other top dead centre. This provides a
perfect axial (walking beam) motion to the tray cells. The motor was driven by the
circuit used earlier for finger wheel. This time, the movement did not damage the cells
and was perfectly aligning the cells to the hole. The walking beam mechanism is
shown in figure 3.5 and plate 3.4.

3.2.4 Growing of vegetable seedling in protray strips for gravity ejection

The protray cell (open on both sides) was kept bottom side up on a flat wooden
plank and filled with conventional growing medium of coir pith (plate 3.5). Single seeds
were sown manually into the media at depth of 5 to 6 mm (Kumar et al., 2009). The trays
were kept in a shaded house and grown. The twenty days old seedlings were tested for
free fall on the test rig developed. Preliminary results showed that most of the plants
did not successfully fall down, since the media stuck to the protray surface and
gravity was not sufficient to release them. To overcome this problem, it was planned
to try different media mixtures and also to use a paper cone within the tray cell, so
that the plugs can release from the cells easily by gravity. Three media compositions
were tried with or without paper cone. The media compositions were coir pith, coir
pith + vermicompost 4:1 by volume, soil and soil + vermicompost 4:1by volume.

Paper cones were made by using chart paper of 1mm thickness so that they fit
exactly into the tapered protray cells. The above media was tried with or without these
cones and a systematic set of trials with five replications each was tried and the data
analysed.

The following problems were observed in the experiment.

1. Though the test rig had a parabolic pathway as slide, the segment carrying the
seedling was quite heavy and slid too fast to the bottom and hit the feeder
mechanism with a force. The subsequent feeds caused them to override the trays
fed previously. This necessitated to add one more intermediate drive to feed the
trays more slowly, which would turn out to be cumbersome and was not done.

2. The segmented tray could not be manually sliced with precision and took some
lateral swing while travelling down the rig causing misalignment.

60
3. Since the seedling has to fall through the cell opening, older seedlings which
have more vegetation get caught on the top of the cell while falling through.

4. Though the paper cone improved the scouring to some extent, watering of tray
wets the paper and makes it ineffective.

Altogether, the ejection by gravity was felt unsatisfactory.

3.2.5 Calculation of the paper cone surface development

The paper cone that was inserted in the cells had,

Top diameter (D) = 32.5 mm

Bottom diameter (d) = 22.5 mm

Vertical height (l) = 35 mm, so as to fit snugly into the

inner surface of the protray cell,

D  d
 
Tan α  
2 
l

Where  is the included half angle of the cone

D  d  3.25  2.25 
   
Tan α   
1  2  2
α  tan
l 3.5

 0.5 
Tan α    α  tan
 1  0.5 
  α 8.132
 3.5   3.5 

D 
Sin    2 
x
 1 
Circumference of top = D =  x 3.25 = 102 mm

Circumference of bottom = D =  x 2.25 = 70.6 mm

 d   2.25   D   3.25 
x 2   2    2   79.5 mm , x1   2    2   114.9 mm,
 sin α   sin 8.13   sin α   sin 8.13 
       

The sheet lay out (surface development) was in the form of a segment of circle
as shown in figure 3.6.The layout is rolled and made into a paper cone.

61
 9
77

140

Idle crank 70 Motor driven crank

Crank pin

Toothed pulley

43
Toothed belt
18
Connecting rod Push rod All dimensions are in mm
Fig. 3.5 Double crank with belt drive
Path of the plunger

62
 Plate 3.4 Walking beam mechanism with belt drive

Plate 3.5 Growing of seedling in open bottom protray segments

63
 102

70.60

50
79.
11
4.9
0

All dimensions are in mm

Fig 3.6 Paper cone

64
3.3 Experiment on seedling ejection by clawing out from the top of the protray cell

Nextly, it was planned to try clawing out the seedling plug from the top of the
tray using an appropriate mechanism. It was felt that the four bar mechanism would
be appropriate to try this method by parking the seedling trays on its side and the
mechanism opposite to the cell. A four bar mechanism was designed to extract the
seedling from the tray cell and release it to fall free to the ground. The tip of the claw
as an extension of the coupler bar was supposed to extract the plug seedling entering
into the top of the protray cell and tease it out, by moving in a predefined path. But
care should be taken not to hit the seedling stem when it was done. A computer
freeware by the name “four bar” (www.gtrebaol.free.fr) was used to synthesise the
coupler curve of the designed mechanism. A path was first assumed to do the job and
the coupler curve analysis using the freeware was done to optimize the dimensions of
the four bar mechanism appropriately to generate the required path of the claw. Figure
3.7 shows the set up used for experimenting this mode of ejection with the inset 3.7. a
showing the actual coupler curve finalized for the mechanism.

3.3.1 Experimental setup for assessing extraction performance of the finalized

mechanism

A tray holder (Fig.3.8) was developed wherein the protray with seedling could
be placed vertically. Three pair of six mm rods with height of 300 mm was mounted
on the holder platform vertically at equal distance with 2 mm gap between the rods
for inserting the protray from above. The tray holder was made to slide horizontally
on a rectangular bar and the whole assembly was fixed on a flat wooden plank.
A handle was fitted at the end of the tray holder for moving the protray forward or
rearward. The protray along with the seedlings could hence be moved and positioned
against the ejecting mechanism, so that ejection performance could be assessed
against each seedling in a row. For assessing the bottom or top row, the holder
platform was raised up on a set of wooden blocks to the required height. The four bar
mechanism was mounted on a vertical stand (Fig. 3.7) with a handle provided for
hand cranking. The handle when rotated in the appropriate direction made the pick -
up pin of the claw enter the top side of the protray cell to a depth of 30 mm, teased
and ejected out the seedling.

65
 Tray cell

Driven crank

Crank rotated by hand

Coupler path

Coupler extension
(claw) DETAIL J

Coupler
Fig 3.7.a Finalized coupler curve of the mechanism Fig 3.7 Setup for clawing out the plug from top of the pro-tray

66
 Tray holder

Handle

Clamp

Slider

Isometric view

Fig 3.8 Tray holder

67
 The pick - up claw was expected not to break up the root ball during removal.
In order to avoid root ball damage, the tips of the claw were flattened. The protray
seedling removal was successful but the root ball being fragile often broke up. This
method of ejection was hence discarded. Plate 3.6 illustrates the construction of the setup.

3.4 Experiments on popping out the seedling from the back of the tray

The third method of ejecting the seedlings out of the protray was contemplated
next, wherein it was planned to insert a plunger from the back of the cell to pop it off
the cell. The existing trays however does not possess a large enough hole to insert a
plunger and hence was modified accordingly.

3.4.1 Experiments on popping out seedling by a hand cranked plunger

A slider crank mechanism was devised which was operated by hand cranking.
The bottom of a conventional 14 x 7 protray was modified with a 10 mm hole at the
cell bottoms. The seedlings were grown in these trays by using conventional growing
medium for twenty five days, for experiments with this method. A 6 mm round rod
was used as a plunger for pushing out the seedling from the pro tray. The rod entered
to a depth of 80 per cent of the protray cell depth so that the seedlings were positively
ejected out. The dimension of the slider crank mechanism was designed to achieve
this, particularly the crank radius. Preliminary trials were found to be satisfactory, so
the ejection of the seedling by popping from back was decided to be motorised.
Figure.3.9 and plate 3.7 shows the details of slider crank mechanism.

3.4.2 Experimental setup for popping the seedling from the back using a
motorized mechanism

3.4.2.1 Protray holder

The protray holder that was developed earlier was used in these experiments
also. The tray holder (Fig.3.8 and plate 3.8) held the pro tray seedling vertically in
sidewise position, ready for ejection.

3.4.2.2 Motorized slider crank mechanism

A 12 V, 100 rpm, DC geared motor was used to operate the slider crank
mechanism. The geared motor drove a crank of 25 mm radius and connected to the
plunger rod sliding in a hollow cylinder mounted on the front similar to the hand
cranked unit.
68
 25 50
100

100

Plunger

168
Isometric view

All dimensions are in mm

Fig. 3.9 Slider crank mechanism

69
 Plate 3.6 Four bar mechanism

Plate 3.7 Slider crank mechanism

70
 The plunger reciprocated through 50 mm and penetrated into the cell bottom to
an extent of 40 mm. An aluminium block with a hole accommodated the plunger’s
sliding action. This assembly of the motorized slider crank mechanism (Fig.3.10 and
plate 3.9) was mounted on a steel plate and a slot was provided vertically throughout
the length. This plate could hence be moved vertically over a fixed rectangular plate
anchored to the table. The slot accommodated a pair of winged nuts and bolts so that
the moving plate along with the plunger could be moved up and down to align it, with
each row of the tray cells. The electrical system was powered by a 12 V battery.

The horizontal movement of the protray was effected by moving it on the tray
holder as explained earlier in section 3.4.2.1. This bi axial movement provided for
aligning the plunger exactly at the back of each cell. The popping out of the seedlings
was to be done from bottom row first upward so that the ejected plant could fall down
freely without entanglement.

3.4.2.3 Regenerative stopping of the ejecting plunger

Actuation of the plunger to eject the seedling plug out of the protray could be
done either pneumatically or electrically. Pneumatic ejection would need a mini
pneumatic cylinder controlled by a solenoid valve. But the major disadvantage is that
a compressed air source is needed on the field machine contemplated. Deriving
compressed air for a field machine is a difficult and costly proposition needing a
compressor, storage tank, FRL, etc., carried and powered by the tractor. Hence it was
decided to go for electrical actuation of the plunger.

Although an electrical solenoid could be used, the following difficulties were

anticipated.

a. The stroke required is about 50 mm since the plunger has to completely eject
the seedling plug out of the 40 mm deep protray cell. This stroke is too large
for a standard solenoid and hence a more powerful solenoid with some form of
mechanical amplification of the stroke was needed.

b. The plunger speed of the solenoid is very high and is difficult to control. Too
high a plunger speed may throw the vegetable plug forcefully too far.

c. A solenoid is less efficient than a motor and consumes lot of power for the
given work.

71
 Tray holder

Pro tray holder

Motor Push button

Cylinderical
Hollow

Plunger
Isometric view

Fig. 3.10 Motorized slider crank mechanism

72
 Plate 3.8 Protray holder

Plate 3.9 Motorized slider crank mechanism

73
 d. DC solenoids are not readily available in the Indian market and are quite
expensive.

Hence it was decided to go for the 12V DC geared motor driven slider crank
mechanism actuating the ejecting plunger as explained earlier. Here the plunger stroke
was easily achievable by having a crank radius of half the stroke (25 mm). The speed
of the DC geared motor (100 or 150 or 200 rpm) is also selectable and can be chosen
to match the velocity of ejection required. 12V DC geared motors are very much
available in the market at relatively cheaper cost of Rs 350/- each. (Chinese make)

But the only problem was that when the plunger motor is operated to move the
plunger forward and retract, switching off the motor after one such stroke may not
instantly stop the motor and so the plunger. Due to inertia, the plunger would move a
bit further and stop at some random position inside the protray cell, thus obstructing
the linear movement of the protray itself for extracting the other cells using the same
plunger. Hence it was conditional that the motor should be stopped instantly after the
plunger makes one stroke ejecting the plug and retracting the plunger back. To
achieve this, the principle of regenerative breaking was used.

Regenerative stopping of an electrical motor is done by way of recovering the

inertial energy of revolution to regenerate electrical energy and then to load it until a
braking effort is put on the revolving shaft. In simpler terms, when a running motor is
disconnected from power, it keeps revolving due to inertia and does not stop abruptly.
But the inertia of revolution of motor generates electricity, which when shorted or
excessively loaded by shorting the motor terminals, the motor armature stops abruptly
due to the excessively heavy electrical load. This is a simple way of stopping a
revolving motor instantly.

The ejecting plunger’s slider crank mechanism is shown in figure 3.10 and plate
3.9, where the crank is rotated by the geared DC motor. An axially adjustable tongue
provided on the plunger axis is strategically arranged to depress a limit switch at the
end of the plunger stroke i.e at the retracted dead centre of the crank mechanism.
The electrical circuit implementing the driving and regenerative stopping of the motor
is shown in figure 3.11. The motor is controlled by a SPDT (Single Pole Double
Throw) relay as shown. The positive supply of the battery is permanently connected
to one terminal of the motor (red line).

74
VCC VCC
GND GND
1 3 3 1 NC
NO NC NO
SW-SPDT
SW-SPDT
2 Limit switch
2 M Motor
M Motor Limit switch
Control signal
Control signal

Relay-SPDT
NC NO
NC NO Relay-SPDT

Fig 3.11. a The relay in normally closed Fig 3.11. b Regenerative stop position
position

VCC VCC
GND GND
3 1 3 1
NO NC NO NC
SW-SPDT
SW-SPDT
2 2
M Limit switch M
Motor Motor Limit switch
Control signal Control signal

Relay-SPDT Relay-SPDT
NC NO NC NO

Fig 3.11.c Relay is switched on Fig 3.11. d The limit switch back to NC
position

Fig 3.11 Regenerative stopping of ejector motor - circuit

75
 In figure 3.11.a, the relay is in normally closed (NC) position, since it has not
been actuated yet (switched off). The limit switch would be in normally closed (NC)
too, since at the beginning of operation of the machine, the plunger would not be at
the end of stroke. So the ground terminal of power is routed through these NC
contacts to the other terminal of the motor (green line). The motor starts to rotate till
the plunger reaches the end of stroke, where the limit switch is depressed and opens it
(Fig. 3.11.b). Now the two motor terminals are shorted together after cutting the
ground off, causing a regenerative stopping of motor instantly at this position. Now
the mechanism stands ready for actuation.

If now the control relay is switched on (Fig.3.11.c), the relay's normally open
(NO) contacts close and direct the ground to the motor's other terminal (green line)
thus rotating the motor. The relay is operated momentarily to move the plunger
forward, releasing the limit switch back to NC position (Fig.3.11.d). Irrespective of
the relay's status now (Fig.3.11.a or d), the motor will continue to revolve until the
plunger retracts back and stops instantly. Thereby, the plunger takes just one stroke
for each momentary switching of the controlling relay. The time of actuating the
SPDT relay should be less than the time of one revolution of motor so that the
stopping is achieved with one stroke.

3.4.3 Experiments on motorized plunger in popping out the plugs

The main purpose of these laboratory experiments was to determine the effect
of influencing variables on the successful removal of the plugs from the protray as
popped from the cell. The four selected independent variables were age of seedling,
growth media, plunger diameter and the speed of plunger. The success of ejection or
failure was recorded in each replication and logit analysis was attempted on the
generated data.

A completely randomized design of statistical experiment was planned, wherein

the response variable is a binomial one as to be linearly modeled (Generalized Linear
Model - GLM) over the four independent variables. The results were expected to
optimize the variables to achieve maximum ejection of plugs.

76
a. Age of seedling

Conventionally, the farmers practise transplanting of 20 days old plug seedling.

But it was planned to test 30 and 40 days old plug seedlings also, since the seedlings
would be easier to eject if the root mass formed is solid by age. Hence the three levels
planned were 20, 30 and 40 days old seedlings. But too much vegetation was expected
to cause tangling of seedlings between each other.

b. Growth media

The vegetable seedling growers mostly use decomposed coir pith alone or coir
pith with vermicompost at a ratio of 4:1 by volume. If the hardness of growth medium
was more than the coir pith, it was felt that the popping and dropping would be better.
Moreover from previous experience the root ball would be less prone to disintegration
while popping, with the use of other media. The straightening of seedling after
ejection was also another criterion to be considered, so that a heavier root ball would
turn upright for placement in the furrow in a shorter height of fall. Accordingly five
treatment media composition (Table 3.1) were used.

c. Plunger diameter

While popping out with the plunger, it should not pierce the plug rather than
pop. A bigger diameter may cause problem in aligning plunger to protray because the
hole at the bottom of the tray is relatively smaller. Hence four treatment plunger sizes
were tried. The plunger tips were made replaceable with these sizes.

d. Plunger speed

The momentum involved in popping action decides proper ejection and

dropping. The momentum if too high would throw the plug far and dropping to
ground would be disturbed. Hence the plunger speed was considered. The horizontal
projectile path of the plug after ejection should be minimized but the momentum
should be sufficient to eject the plug. Three treatment speeds (Table 3.1) were tried.

77
Table 3.1 List of selected variables for popping out the seedling from protray

Variables Levels
Independent
Age of seedling (A) 20,30 and 40 days old seedling - A1, A2 and A3
Coir pith (M1)
Coir pith + vermicompost (4:1)(M2)
Coir pith + soil (1:1) (M3)
Growth media (M)
Clay soil + sand + coir pith (1:1:1) (M4)
Soil + sand + coir pith (2:1:1) (M5)
- M1, M2, M3, M4 and M5
Plunger diameter (D) 3,4, 5 and 6 mm - D1, D2, D3, and D4
Plunger speed (S) 100, 200 and 300 rpm - S1, S2 and S3
Dependent or response variable
Binomial - success or failure
No of replications in
R1, R2 and R3
each treatment

The results of this experiment was expected to optimize these parameters for
maximum successes of ejection.

3.5 Optimization of agronomical parameters of the growth media

The seedling ejection from the protray alone is not adequate, but the plant itself
should grow healthily for transplanting in the selected medium. Inadvertently
designing the media for machine suitability alone would not be acceptable. So to
assess the plant growth as influenced by age and growth medium, they were
experimented in terms of plant height, stem diameter, number of leaves, plug weight,
root mass and stem weight.

3.5.1 Construction of shaded net house for growing of protray seedling

A shaded net house of a 3000 x 1500 x 1500 mm (plate 3.10 and plate 3.10. a)
was exclusively developed for raising plug transplants with different growth medium.
A cuboidal metal frame was built over which shaded net was wrapped over. A gate
with shaded net was provided on the side for entry and exist. The 75 per cent shade
net was used to cover the shade house to protect the seedling from direct sun light.

78
 On both sides of the green house floor, 300 mm high flat soil bed was formed
to prevent water stagnation and it was fully covered with 1 mm thick black polythene
paper to avoid root contact between the soil and plug tray through the drain hole.

The main difficulty faced in upkeep and growing of the protray seedling was
due to the provision of large hole at the bottom of the cells. There was no problem in
keeping the media in because it is compacted into the cells before sowing. However
when these trays were placed on a flat plastic surface, the water draining and
stagnating as a film beneath promoted growth of roots out of the bottom of the tray.
These roots would impede proper ejection afterwards.

Moreover, when more number of protrays were demanded for experiments,

help of a local protray grower was sought, who grew the seedling in the desired media
for experimentation. The grower since of the mass production was not, willing to
grow the seedlings in our modified protrays. He inserted these trays into the
conventional trays with very small drain holes, then filled media and grew the
seedlings. Here also the problem of roots growing out of the modified tray was
observed, which for experimental purpose were trimmed off. But it must be noted that
protrays used in the western countries had very large drain/pop holes, similar to our
modified trays. However when the media loaded trays are placed in racks with their
bottoms open to air, the roots does not propagate out of the tray.

3.5.2 Rack system

A rack system hence was developed for growing the plug transplants in the
constructed shaded house. MS angle (25 x 25 x 3 mm) was used for fabrication of the
rack and 6 mm round rods were fixed lengthwise supporting the protrays. The rack
had four partitions each holding four trays, totally accommodating 12 trays at a time.
The seeded trays were placed on the racks and each tray must be handled individually
to load and unload from each shelf throughout the growing process. The detail of the
developed rack system is shown in figure 3.12 and plate 3.11.

3.5.3 Experiments on the agronomical status of the seedling in various media

A planned statistical experiment was executed to assess the agronomical factors

influencing the growth of seedlings. The independent variables and the levels of
treatments thereof are shown in table 3.2. The analysis of the result was expected to

79
show whether the growth is affected by the change in media and age. The media and
age could be optimized for better performance on the planned machine.

Stem diameter was measured in mm using a digital vernier calliper, The height
of the plant was recorded with a steel foot rule. In addition to the agronomic health of
the plants, the root plug weight and stem weight were also measured and analysed in
the experiment. The plug weight was measured after the bottom portion of the
seedling was removed from the protray manually. The upper portion of the seedling
was removed and the plug (root ball) was weighed. The top portion of the seedling
from the tray cavity (root ball) as already removed was measured as stem weight in
grams. It was expected to see whether the media and age influences the root mass
weight which may be crucial for proper falling of plug to the furrow. The seedling
was taken out manually from the tray and the root ball was removed from the stem.
The growth medium was washed off and wet weight of root was recorded.

Table 3.2 List of selected variables for agronomical parameters

Variables Levels
Independent
Age of seedling (A) 20,30 and 40 days old seedling - A1, A2 and A3
Coir pith (M1)
Coir pith + vermicompost (4:1)(M2)
Coir pith + soil (1:1) (M3)
Growth media (M)
Clay soil + sand + coir pith (1:1:1) (M4)
Soil + sand + coir pith (2:1:1) (M5)
- M1, M2, M3, M4 and M5
Plant height
Stem diameter
Number of leaves
Dependent or response
Root weight
variable
Plug weight
Stem weight
Pulling force
No of replications in each
R1, R2 and R3
treatment

80
3.6 Pulling force measurement

The pulling force to remove the seedling from the tray was measured to find any
variation in ejecting out the seedling form the protray in terms of growth medium and
age of seedling which may affect the ejection of seedling. It was measured by using a
spring balance. The spring balance is fixed on a arrangement which is provided to
move up the balance holder and down while rotating the handle and the pulling force
was measured directly from the scale. The details are shown in figure 3.13 and plate
3.12.The measured pulling force as influenced by the media and age were analyzed
along with the responses as shown in table 3.2.

3.7 Drop test for weight loss analysis of root ball and straightening height

The ejected seedling may fall down to the ground directly or on a trap door
during transplanting, which may cause damage to the root ball. To determine the
weight loss of root ball of the optimized medium while falling down, the drop test was
conducted at different heights of 1000, 750, 500 and 250 mm. For conducting this
experiment a wooden board was selected at 1000 mm height x 1000 mm width and
marked at different height at an interval of 250 mm. The seedling was dropped
manually from different heights. The seedling used for testing is either without
hardening (The seedling was irrigated before one day of transplanting) one day, one
and half day and two day hardening. Plate 3.13 shows the details of drop test setup.

3.7.1 Binding materials for minimizing the effect of root disintegration

The growth media was found to disintegrate during the experiments conducted
on plant ejection. So the drop test was done to find the height, which may avoid root
ball disintegration while falling through a height. Attempt was made to reduce the
root ball disintegration by adding some binding materials which may firm the growth
media with roots and reduce the root ball loss. Binding materials such as cow dung,
neem gum, synthetic gum and clay soil was added to the growth media and seeds
sown in the tray. After 20 days the seedling was pulled out manually from the protray.
But while removing the seedling from the protray, primary trials showed that the clay
soil mixed growth media was only removed easily without being fragile while other
binding materials stick to the protray. Plate 3.14 shows the protray seedling with
binding materials.

81
835

200

400 1000

Isometric view

Fig 3.12 Rack for growing of seedling

Handle

Spring balance

Tray

All dimensions are in mm

Fig 3.13 Pulling force measurement assembly

82
Plate 3.10 Shaded net house exterior view

Plate 3.10.a Shaded net house inside view

Plate 3.11 Rack with protray seedling

83
Plate 3.12 Pulling force measurement assembly

Plate 3.13 Experimental set - up for drop test

84
 Nextly clay soil was sieved and dried in open sun for 30 h continuously for
removing pathogens and mixed with coir pith at 2, 5 and 10 per cent by volume. The
seeds were sown and the seedling was compared with conventional medium (coir
pith) grown seedling. It is observed that there is no significant difference between coir
pith, coir pith + 2 per cent clay soil and coir pith + 5 per cent clay soil in terms of
number of leaves, stem diameter and plant height. The germination percentage was
found to be very less in the coir pith + 10 per cent clay soil growth medium. So these
two media were selected for further experimentation.

The optimized medium with binding materials (clay soil 2 per cent + coir pith,
clay soil 5 per cent + coir pith) was selected for conducting drop test. The results
showed that the weight loss of the root ball was reduced to some extent due to adding
binding materials but not up to the expected level. Hence the provision of cushion was
also attempted in reducing the weight loss of the root ball.

3.7.2 Weight loss with cushion

The seedling was dropped down manually on surfaces with a cushion of

20 and 40 mm thickness. The sponge was used as cushioning material for the dropped
seedling. Table 3.3 shows the details of the factors that affect the root ball weight loss
while falling down through a height on a surface as influenced by the media with
binding as well as hardening. A standardized experiment with the shown factors was
systematically conducted to optimize the factors within the levels indicated so as to
use these in the prototype.

3.7.3 Straightening height of the seedling

To find the straightening height of the seedling which is used for designing the
hopper in the vegetable transplanter, the motorised slider crank mechanism was again
used to eject the seedling as explained in the section 3.4.2.2. A hard board cone was
fixed on a rectangular frame through which the seedling will fall down and
arrangement was provided to fix the frame at different heights of 250, 500 and 750
mm. The ejected seedling was collected at the bottom of the cone on a surface. The
number of plants straightening up successfully was noted and analysed. Plate 3.15
shows the experimental set - up. Table 3.4 shows the factors and their levels selected
for the experiment.

85
Table 3.3 List of selected variables for root ball weight loss analysis

Variables Levels

Independent

Coir pith (M1)

Growth media (M) Coir pith + Clay soil 2 per cent (volume) (M2)
Coir pith + Clay soil 5 per cent (volume) (M3)

Height (H) 500 mm, 750 mm and 1000 mm, H1, H2, H3

Without hardening,1day,1.5 and 2day after

Hardening(D) (Day)
D1, D2 , D3 and D4

Without cushioning , 20 and 40 mm C1, C2 and

Cushioning (C)
C3

Dependent or response
Percentage of weight loss
variable

No of replications in each
R1, R2 and R3
treatment

Table 3.4 List of selected variables for straightening height

Variables Levels

Independent

Coir pith (M1)

Growth media (M) Coir pith + Clay soil 2 percent (volume) (M2)
Coir pith+ Clay soil 5 percent (volume) (M3)

Height (H) 250, 500 and 750 mm H1 , H2 and H3

Dependent or response
Number of plants falling down straightly
variable

No of replications in each
R1, R2 and R3
treatment

A logit analysis was done on the number of successful straightening of

seedlings as influenced by the independent variables to ascertain the optimum height
of drop for successful straightening.

86
 Plate 3.14 The tray seedling with binding materials

Plate 3.15 Experimental set up for measuring plant straightening height

87
3.8 Development of a prototype vegetable transplanter

The prototype vegetable transplanter was developed based on the optimized

variables obtained from the experimental data. The prototype consisted of a main
frame, drop hopper, furrow opener and compaction wheel.

3.8.1 Main frame

The main frame was used for fixing the assembly of the furrow opener, hopper,
and compaction wheel. It is attached to the tractor three point linkages. The overall
dimensions of the main frame are 1290 mm length, 810 mm width and 460 mm
height. The different views of the frame are shown in figure 3.14.

3.8.2 Hopper

The hopper was designed to drop the seedling to the ground without any
interference. The bottom of the hopper was closed by a spring mounted hinged trap
door, which remains closed until the seedling is released to the furrow and again the
trap door closes after its passage. The seedling would stay on the trap door before it is
released to the furrow. The opening and closing of the trap door is operated by a cable
which is attached to a ground wheel. The cable is operated by a separate tripping
arrangement. When the ground wheel hub rotates, a 5 mm round rod as inserted in the
hub activates the flap door through the cable and retraction done by a spring. The trap
door is opened and closed at every 450 mm spacing on the row accordingly. This
allowed readily positioned seedling to slide into the furrow. After the seedling was
planted, the flap door closes and the next seedling falls onto it in position to complete
the planting operation. Figure 3.15 shows the details of the hopper and flap door.

3.8.3 Furrow opener

A furrow opener was developed to open a narrow furrow for a width of 60 mm.
It was mounted on 100 x 50 x 12 mm (L x B x D) rectangular drop. Holes were made
on the drop structure every 25 mm interval for fixing the desired depth. Side covers
were fixed on the furrow opener for preventing soil flow into the valve mechanism
while opening a furrow in the field. Figure 3.16 and plate 3.16 shows the details of
furrow opener.

88
 Main frame Protection clamp

460
1290 880

810
400 Isometric view

All dimensions are in mm

Fig 3.14 Main frame

89
3.8.4 Compaction wheel

The adjustment of the compaction wheel and forward speed is important factors
for proper coverage and compaction of the transplanted seedling. 150 mm diameter
solid rubber wheels were used as compaction wheels and were held at the back of the
furrow opener. The wheels had provision to adjust their roll angle as well as their yaw
angle. The roll and yaw angles are vertical and horizontal angles between the plane of
the wheel and the vertical plane along travel direction. These are provided to collect
the soil and compact it around the planted seedling. Figure 3.17 and plate 3.17 shows
the details of compaction wheels. The seedlings fall just before the wheel position so
that compaction is possible.

3.9 Field level optimization of variables for the vegetable transplanter

The main purpose of the field optimization was to determine the influence of the
transplanting variables on planting performance. The four selected variables were
speed of travel, planting depth, compaction wheel angle (yaw) and soil moisture
content as influencing the field performance of the vegetable transplanter (Table 3.5).

3.9.1 Speed of travel

Travel speed of the planter influences the time available for the mechanism to
plant, each seedling. For instance, a transplanter travelling at 1 km h -1 i.e. 270 mm s -1
speed, will have only 1.62 s to plant one seedling at 450 mm spacing. Within this
time, the seedling has to be ejected, fall to the check valve, be retained, dropped into
the furrow and then closed and compacted. Taking into account all the mechanical
lags, time of fall, etc, the speed would hence be a limiting factor on the performance
of the machine. Hence three levels of forward speed were selected viz ; 0.5, 1.0 and
1.5 km h-1.

3.9.2 Planting depth

The depth of planting affects proper establishment of the plant and thus
performance of the transplanter. If the planting depth is not optimal, the seedling will
not stand straight in the furrow. This may lead to toppling and burial of the seedling in
the furrow. Three levels of planting depth. viz; 50, 75 and 100 mm were considered
for optimization.

90
 168
300

125

851

81
° Isometric view
64
(a) Hinged check valve

Fig 3.15 Hopper

Soil shield
600

26
°
145

Share
68
473 Isometric view
215
All dimensions are in mm

Fig 3.16 Furrow opener

91
425

Rubber solid wheel

520

Isometric view

60
300 All dimensions are in mm

Fig 3.17 Compaction wheel

92
 Plate 3.16 Furrow opener

Plate 3.17 Compaction wheels

93
3.9.3 Compaction wheel angle

The compaction wheel angle (yaw) is important for making the soil firm around
the transplanted seedling. The roll angle is kept constant at about 20  to the vertical.
The angles may affect the soil flowing into the furrow as large volume of soil may
flow through the compaction wheel and the plant will be buried in the furrow. If the
angle is low, insufficient soil flows into furrow and it is not enough to firm around the
plant. So three different compaction wheel angle was selected for optimization viz., 0,
15 and 30 .

3.9.4 Soil moisture content

Soil moisture is the most important factor affecting the performance of the
vegetable transplanter. If the soil is loose and dry, it flows in too fast around the root
ball before the plant properly orients itself in the furrow and if the soil is too wet,
sufficient soil will not flow around the root ball. Hence 10, 15 and 20 per cent (w.b)
soil moisture content were selected for field level optimization.

The field was irrigated and allowed to dry out to approximately18 to 20 per
cent moisture content and then rotovated twice for proper mixing and pulverization.
Random soil samples were taken and soil allowed to dry to attain the desired level of
soil moisture. Transplanting was first done at 20 per cent moisture content. The
transplanting was then carried out after 2 - 3 day interval as the soil moisture reached
to the desired levels of 15 per cent and 10 per cent.

Table 3.5 List of selected variables for field level optimization by manual
dropping

Variables Levels
Independent
Speed 0.5, 1.0 and 1.5 km h-1 , S1, S2 and S3
Furrow depth 50, 75 and 100 mm D1 , D2 and D3
Compaction wheel angle 0, 15 and 30 C1, C2 and C3
10, 15 and 20 per cent (approximately) M1, M2 and
Soil moisture
M3
Dependent or response
Plant uprightness
variable
No of replications in each
R1, R2 and R3
treatment

94
 Logit analysis was done on the success of plant uprightness so as to optimize
the variable levels. The seedlings were manually dropped from a height of 500 mm
into the hopper at regular intervals of 2 seconds so as to test the performance. Since it
was manual method, the spacing was not taken as response.

3.10 Development of the automatic transplanter mechanism

Having optimized the field performance parameters of the prototype

transplanter, the automatic transplanter mechanism was devised next to automatically
pop out the seedlings and plant them in the furrow.

The automatic transplanting mechanism had the following constituent parts.

1. A horizontal rail and slider assembly carrying the vertically loaded seedling
protray, which slides along the travel direction of the machine

2. A vertical column of seven ejector mechanisms that are positioned so as to

eject plugs from one single vertical column of the protray at a time

3. A motorized walking beam mechanism to move the tray slider in precise steps
corresponding to the inter distance between cells on the protray.

4. Optical detector for precise positioning of the pro-tray cells relative to the
ejector plungers.

5. Main microcontroller board controlling the logic of tray movement and plug
ejections in proper sequence.

6. An exclusive relay board for driving the seven ejector motors along with
hardware for regenerative stopping of these motors.

The protray slider assembly (Fig.3.18, a) is made of extruded aluminium

sections to make it light in construction so that the effort for moving the pro-tray is
minimized. The main slider is hollow aluminium square of 40 x 40 mm size, which
slides in a 50 mm aluminium open channel of 1000 mm length (Fig. 3.19,a). The
slider is placed on a series of rollers (Fig. 3.19, b) that are positioned at the bottom of
the channel section. The rollers used for sliding windows were used as available off-
the-shelf. Three such sets of parallel rollers carry the whole length of the slider box.
Similarly the sides of the channel into which the slider sits in, was also provided with
these rollers three in series on one side and another set of three on the other side

95
(Fig.3.19,c). To restrain the slider from the top, suitable rollers are positioned on the
front and back of the channel (Fig.3.20, a) such that they press the slider down into
the channel. With this arrangement, the slider box is completely restrained from all
the sides inside the channel section, but can slide on these rollers axially within the
channel rail. The roller assemblies were adjusted precisely for the right clearance and
the slider was provided a frictionless movement axially. A set of forks of height
(Fig.3.18, b) corresponding to the protray width were mounted on the slider box so
that a protray with seedlings could be inserted down between the forks. The protray is
to be inserted into this arrangement such that the bottom of the protray faces the
ejector motor assembly stationed at the back of the slider assembly.

The walking beam mechanism is one which is intended to provide an

intermittent precise linear movement to the slider by converting a rotary motion to a
linear movement. This mechanism was selected over a typical screw and nut assembly
to make it simple. The mechanism as already explained in section 3.2.3 consists of
two crank wheels of equal crank radius. It is a double crank arrangement where the
crank pins of the two cranks are connected by a connecting bar. This arrangement is
just a four bar mechanism with equal crank lengths. The two cranks are actually
arranged on a two toothed belt pulleys of the same diameter of 20 mm. A toothed belt
of size 100 mm and with teeth of 1 mm pitch is used over these pulleys. One of the
pulleys is the driving pulley and is coupled to a 12V DC 100 rpm geared motor
mounted vertically beneath the double crank assembly (Fig.3.5). The pulley when
driven by the motor revolves the other pulley at the same angular displacement, which
is mounted as an idle pulley on a stub axis. The coupler bar as driven by the double
crank hence makes circular motion whose axis always remains perfectly parallel to
the line connecting the two crank centres. The coupler bar in turn has an extension
vertical to the axis of the bar and in plane with the line of motion. The coupler point
on the tip of this extension has a circular motion of 36 mm diameter. An object kept in
its path but restrained to move only linearly parallel to the line connecting the crank
centres, will be pushed exactly 36 mm after which the coupler’s extension end
releases the object. This mechanism is mounted on a steel pedestal in line with the
slider assembly (Fig.3.20, b).

The slider box, carrying the protray is provided with a lateral extension
(Fig. 3.20,c) on the front which carries a set of 14 nylon buttons of 35 mm diameter

96
each, arranged in a line parallel to the slider movement (Fig.3.18,c). This provision
(plate 3.18 and 3.19) integral to the slider assembly slides along with it. The walking
beam mechanism mounted on a suitable pedestal is aligned to this button rack in such
a way that the coupler’s extension can travel into the gap between the buttons and
push each button in turn. When the drive motor of the walking beam mechanism is
driven one revolution, the button rack along with the protray in turn is moved through
36 mm linearly , which is the inter cell distance of the pro-tray.

The tray drive motor (12 V, DC geared,100 rpm) is not directly driven but
through a regenerative stop circuit (Fig.3.11). The relay contacts see that the motor
terminals are shorted together when the motor is switched off so that regenerative
stopping is achieved. This provides for stopping the protray movement exactly at the
required spacing so that the ejector mechanism is aligned to the centre of the protray cell.

The ejector assembly (Fig.3.21 and plate 3.20) has a vertical array of seven
ejector mechanisms arranged one over the other on a vertical plate stand. As
explained already in section.3.4.3, each ejector assembly has a plunger (Fig.3.21,a)
(slider) driven by a 12V (Fig.3.21,b) geared motor driven crank (Fig.3.21,c) of 20 mm
radius. Each assembly has its own limit switch arranged at the front to implement
regenerative stopping as explained in section 3.4.2.3. The seven motors are mounted
from the back of the vertical mount plate in a staggered fashion to accommodate for
the vertical spacing of 36 mm between the plunger tips of the seven assemblies. This
is since the crank diameter being 40 mm should not intervene with the motion of the
adjacent crank. The relay drive board intended to drive these motors in sequence
from bottom to top is mounted on the vertical mount plate and the motors and
switches wired appropriately from it. The PCB (Printer Circuit Board) of the relay
board was designed and fabricated (Fig.3.22 and plate 3.23). The vertical mount plate
is fitted on a support frame (Fig.3.21, d) and has provision for adjustments in three
axes, so that the plunger tips can be aligned to the exact centres of the protray cells.
The spacing between the seven sets of plungers are precisely same as the inter cell
distance of the trays. As already explained in section 3.4, the plunger ejects the
seedling plug from the back of the protray, whose cell bottoms are open for such
operation. When one column of protray cells are positioned in alignment with the
plunger assembly, the plugs from the bottom up will be ejected one by one from all

97
the seven cells in that column of protray. Then the slider/protray is moved to the next
column and the ejection is repeated.

Although the walking beam mechanism provides correct linear movement of the
protray by 36 mm towards aligning the centre of cells to the plunger assembly, it was
not precise enough. Hence an optical detector was added to the arrangement. The tray
slider was mounted (Fig.3.20) with a 2 mm thick vertical aluminium strip (interrupter
plate) with 14 x 3 mm equispaced holes at 38 mm spacing (Fig.3.20, d). An opto
interrupter MOI8 was appropriately positioned straddling the holed plate such that the
plate interrupts the opto device (Fig.3.20, e). The sensor is mounted on a suitable steel
holder fitted to the frame (Plate 3.22) with the sensor straddling the interrupter plate.
The opto interrupter board is a small 25 x 25 mm PCB board, which has all the
dropping resistances for the photo diode and the photo sensor of the opto interrupter
circuit. Whenever a hole aligns with the opto interrupter, it detects it and this signal is
connected to the control device. This optical detector provides for a feed back control
on the precise alignment of the pro-tray cells to the plunger assemblies.

The opto interrupter has an infra red led on one side and an infra red detector on
the other. Whenever the emission from the diode falls on the sensor its voltage drop
falls to a low or near zero, since it is in series with a drop resistor. When the emission
is interrupted, the voltage returns to high. A switched signal results that is used for
precise stopping of protray in the required spacing.

The controller board (Fig.3.23 and plate 3.21) which provides the logic for the
sequence of operations of the machine, consists of the main microcontroller chip
16F628A, two numbers of UL2803 Darlington driver ICs and other allied hardware. It
also has a three terminal voltage regulator for converting 12V to 5V for powering the
microcontroller and some noise reduction capacitors along with high frequency filter
capacitors. The 2803 driver ICs are for directly driving the 12V relays of the seven
plug ejector motors, the protray movement motor as well to reverse the polarity of
motor to change directions of tray movement. The DPDT (Double Pole Double
Throw) relay is wired to invert the power connections to the tray movement motor,
when the reversing signal is received from the microcontroller. The SPDT (Single
Pole Double Throw) relay connected across the tray movement motor is for the
purpose of regenerative stopping of the motor. When this relay is activated by the
microcontroller through the A0 bit of port A, the power connections as routed through

98
the DPDT reversing relay reaches the tray movement motor driving it in the required
direction. But when it is switched off, the relay’s NC terminal shorts the motor
terminals together after disconnecting one power wire, effecting an instant stopping of
the motor. This aids in the process of exact alignment of the tray cells to the ejector
plungers. The port B of the microcontroller is dedicated to the ejector motor
activation. The 7 bits from LSB viz, B0 to B6 is used for driving the seven motors.
When a binary data of 80H i.e ‘10000000b’ is loaded into the accumulator of the
microcontroller, right shifted and output to the port B, then the corresponding motor
connected to B6 bit of the B port is activated since output is ‘01000000b’ i.e 40H.
Next right shift makes it 20H, activating motor connected to B5 bit of port B and so
on. The microcontroller accepts a switched signal from the ground wheel for detecting
the plant spacing on the row as well the signal from the opto sensor for precisely
positioning of the pro-tray cell centre with respect to the ejector plungers. (Fig.3.21).
The switch debouncing is done completely by software. The PCB of the main
controller board was fabricated and populated for use.

The equipment is provided with a trailed ground wheel which serves two
purposes namely,

 Sensing the distance moved and issuing a signal corresponding to the desired plant
spacing

 Operating a planting check valve at the end of the drop plenum of the machine at
the desired plant spacing.

The ground wheel is a lugged steel wheel mounted on a fork like frame hinged
to the machine chassis. The wheel with a 300 mm effective diameter is mounted on an
axle supported by a set of bushed bearings. The frame has an appropriately mounted
limit switch, which gets activated by (Fig.3.24, a) a two lobed cam plate (Fig.3.24, c)
mounted concentric to the wheel axle. The circumference of wheel being 900 mm
activates the switch twice every revolution i.e at every 450 mm (desired plant
spacing) travel on the row. This signal from the switch informs the microcontroller to
eject and drop the plugs from the protray. The wheel also carries another two lobed
cam (Fig.3.24, c and plate 3.24), which mechanically actuates a spring loaded check
valve provided at the end of a hopper and drop plenum of the machine as explained
already in section 3.8.2 (Fig.3.15 and plate 3.25) and the hopper plenum (Fig.3.15 and

99
plate 3.25) is provided to receive the ejected seedling without any interference while
falling down. The dimensions are so designed that the ejected seedling falls into the
top opening in spite of its lateral travel. The bottom of the hopper is closed by a spring
mounted hinged trap door (check valve) which remains closed until the seedling is
released in to the furrow and again the trap door closed back. The seedling would stay
on the check valve (Fig.3.15 and plate 3.25) before it is released to the furrow. The
opening and closing of the check valve is operated through a cable which is attached
to toggle drive activated by a ground wheel (Fig.3.24, b). Since the cam and follower
of the valve actuating mechanism is mounted on a hinged support arm of the wheel,
the motion is transmitted to the check valve by way of a cable. The outer sleeve of the
cable is locked on to the frame and the inner one transmits the motion to the check
valve. The two cams are offset by about 30  angularly, so that the plug ejection
precedes the opening of the check valve operation. In simpler terms, the plug is first
dropped into the drop plenum, held by the check valve and then dropped into the
opened furrow. The check valve is lined with a sponged plastic lining, so that the plug
does not get damaged when it falls through drop height on to the valve surface. The
check valve opens enough to slide the seedling plug out into the furrow.

The ground wheel is mounted on the frame of the machine and is designed to
operate the trap door and the micro switch as explained already. The lugs are fixed on
the periphery of the ground wheel at equal distance for better interaction with the soil.
A chain retains the wheel frame not to swing down while the frame is lifted out of
ground. On the ground wheel frame, a dead weight is added to improve the torque
developed by the ground wheel.

The flow of the program as shown in figures 3.25 and 3.26 was hand coded and
then flashed on to the microcontroller chip for operation. As soon as the circuit is
switched on, PORT B of the microcontroller i.e. B6-B0 bits of the port are set as
output to drive the seven DC ejector motor relays meant for one column of protray
cells. PORT A.0 is set to drive the relay controlling the tray movement motor,
PORT A.1 to drive the tray movement motor’s reversing relay, PORT A.3 is set to
input the signal from the switch on the ground wheel detecting the spacing on the row,
and PORT A.2 to input the opto sensor input detecting the tray’s precise position.
Then the program waits for a long press of the manual pushbutton. Once this is given,
it moves and positions the protray to the first column of cells ready for operation.

100
Further signals as obtained from the ground wheel switch detecting the plant spacing,
activates the ejector motors from bottom up sequence. The ejected plugs fall into the
drop plenum, turns upright by gravity and falls straight on to the check valve at the
bottom of plenum. The fallen plug is then released into the opened furrow by the
check valve to precisely fall in the right place on the furrow. As all the seven cells in a
vertical column of the protray are ejected completely, the slider drive motor moves it
to the next column of cells for ejection. This cycle of events is repeated until the
whole protray is emptied out planting 98 plants in a row at the desired spacing. A
counter is initiated and each time one column of tray is emptied, it is incremented till
it reaches to 14, indicating that the tray is empty. Once this is over, the machine needs
to be stopped and the empty tray is to be replaced with a loaded seedling tray.
At this point, the reverse motion of the tray is invoked automatically by the controller,
by reversing the polarity of the tray motor through the reversing relay. Further
switching from the ground wheel, starts the cycles of ejection once again till this tray
is emptied and so on. But this time the tray moves in the reverse direction till it is
emptied. The assembly explained so far was developed first and tested in the
laboratory. A manually operated toggle switch is provided for making the bottom of
the pro- tray align to the plunger initially and the toggle switch is connected to ground
wheel limit switch in parallel. While operating in the field the ground wheel limit
switch activated for continuous transplanting operation.

3. 11 Development of the tractor mounted automatic vegetable transplanter

Having completed the automatic transplanter mechanism as explained in section

3.10, it was integrated to the vegetable transplanter prototype already explained in
section 3.9 to make it into a fully automatic vegetable transplanter (Fig.3.27 and plate
3.26).

3.12 Field evaluation of the automatic vegetable transplanter

The field was prepared by rototilling it twice with a soil moisture sprinkled to
about 10 per cent wet basis. The prototype machine was operated at a speed of
0.5 km h-1, 10 per cent moisture content at 75 mm planting depth with compaction
wheel angle set at 15 . The machine was evaluated for planting at spacing of
450 mm. The plant stand was calculated in terms of number of plants in upright
position and number of plants buried. The planting angle was measured as the angle

101
of inclination of the plant with the vertical. The plants having planting angles of 0 to
30 were considered as likely to produce upright plants. The plants angled at less than
30 with the horizontal plane were considered buried. 20 days old tomato seedlings
with the height of 120 to 140 mm were used for field evaluation. The field after
transplanting is shown in plate 3.27.

102
 115 Protray holder (b)
5

25
250
Square hollow
slider (a)
36

850
900 36

Array of nylon
40
roller (c)

All dimensions are in mm

Fig 3.18 Pro - tray slider assembly

Channel (a)
side roller (c)
41.5

800

Bottom roller (b)

Isometric view

Fig 3.19 Slider

103
 optical interrupter plate (d)

Top roller (a)

Array of nylon
buttons (c)

Opto interrupter
mount (e)
Isometric view

Walking beam mechanism (b) All dimensions are in mm

Fig 3.20 Assembly of the automatic ejection mechanism

104
 Plunger rod (a)
Connecting
rod (d)
36

Crank (c)

268
40
Motor (b)

Isometric view

Support frame (e)

All dimensions are in mm

Fig 3.21 Plant ejector assembly

105
Plate 3.18 Tray slider assembly

Plate 3.19 Slider

106
 Plate 3.20 Plant ejector assembly

Plate 3.21 Main microcontroller board

107
 Plate 3.22 Opto sensor

Plate 3.23 Relay board assembly

108
1
0
9

RLY-COM to MOT2
RLY-coil1 (12V) to SW-NO & MOT1
RLY-NO (-12V) to SW-NC

RLY-NC to SW-COM

RLY-coil2 to board

12V -12V
Fig 3.22 Ejector motor drive relay board

109
Fig 3.23 Main microcontroller board

110
 Dead weight

400
610

Lugs

Limit switch (a)

Check valve
operating cam (b)

Limit switch - lobes (c)

Isometric view
130 50
All dimensions are in mm

Fig 3.24 Ground wheel assembly

111
Plate 3.24 Ground wheel

Plate 3.25 Hopper

112
 Start

Set up a variable to count 14 columns of tray

Set PORT B6-B0 as to drive seven DC ejector
motors on one column of protray cells
Set PORTA.0 to tray movement relay
PORTA.1 to tray movement reversing relay
PORTA.3 to input signal from
ground wheel corres to spacing
PORTA.2 to input optosensor detecting
tray movement position

Initiate ejector motor output bus as binary 1000000

Initiate column count as 1
Switch off all tray movement relays

Start switch activated?

(With debouncing routine)

Rotate tray movement and

look for opto sensor position detect

D A

Fig 3.25 Flow diagram Continued .....

113
D A

First column
position detected?
N

Stop tray movement and

enter the ejection sequence

Spacing switch on
ground wheel activated?
N

Right shift the ejector

output to eject cells
from bottom to top.
B
N All seven cells ejected? Y

Fig 3.25 Flow diagram Continued ............

114
 B

Rotate tray movement and

look for opto sensor position detect

Next column
position detected?

Increment column count

Is the column count
Y 14?
Y
N

Invoke reverse
tray movement.
Initiate ejector motor
output bus as
Initiate ejector motor
binary 1000000
output bus as
Initiate column
binary 1000000
count as 1.
Change protray

Fig 3.25 Flow diagram

115
 12V
GND
12V VR1 5V
Vin Vout
Ejection motor drive circuit (x 7)
C1
Cap Pol1 GND S1
1000mF Limit switch SPDT
5V 7805 GND
GND P1
GND U2
R1 1
P2 M1 1 18 M3
10K M2 2
IN1 OUT1
17
2 M Motor
1 10K IN2 OUT2 3
M3 3 16
2 IN3 OUT3 4 12V
M4 4 15
IN4 OUT4 5
M5 5 14
Header 2 MaleGND IN5 OUT5 6
M6 6 13 K1
IN6 OUT6 7 Relay-SPDT
M7 7 12
IN7 OUT7 8
5V U1 MM 8 11
IN8 OUT8 9
14 17 MM
VDD RA0/AN0 10
18 MR 9 10
RA1/AN1 GND COM D 11
5 1
VSS RA2/AN2/VREF 12
2 P2
RA3/AN3/CMP1 12V 1 GND
RA4/T0CKI/CMP2
3 GND ULQ2803A 2
GND
RA5/MCLR/VPP
4 Header 12 (Male) 3
RA6/OSC2/CLKOUT
15 U3
4
Ejection motor
RA7/OSC1/CLKIN
16 MR 1
IN1 OUT1
18
5
outputs (x 7)
2 17
IN2 OUT2 6
6 M7 3 16
RB0/INT IN3 OUT3 7
7 M6 4 15
RB1/RX/DT IN4 OUT4 8
8 M5 5 14
RB2/TX/CK IN5 OUT5 9
9 M4 6 13
RB3/CCP1 IN6 OUT6 10
10 M3 7 12
RB4/PGM IN7 OUT7 11
11 M2 8 11
RB5 IN8 OUT8 12 12V
12 M1
RB6/T1OSO/T1CKI/PGC
13 9 10
RB7/T1OSI/PGD GND COM D
Header 12 (Female)
PIC16F628A-E/P GND ULQ2803A

12V
K2
M8
M
K?
GND
5V
Tray movement Motor
R11 R12
1K
U? 1K
1K 1K 12V 12V
Relay-DPDT Relay-SPDT

Optoisolator1
GND GND

GW actvated
Selecter S1 Fig 3.26 Schematic diagram for working of the prototype automatic
S5
P2
2
3 vegetable transplanter
1
1
2 S1

Header 2 Female
Manual inch
116
 Hopper
Main frame Plant ejection
assembly
Protection clamp
1017

Furrow opener Ground wheel

Compaction wheel 1064

1449

Slider
assembly
Isometric view
All dimensions are in mm
Fig.3.27 Prototype automatic vegetable transplanter

117
Plate 3.26 Prototype automatic vegetable transplanter

Plate 3.27 Transplanted field

118
Results and Discussion
 CHAPTER IV

RESULTS AND DISCUSSION

A tractor operated automatic vegetable transplanter was designed, fabricated and

evaluated in the field. The results of the field investigations, laboratory studies, design
and development process, and performance evaluation of the automatic vegetable
transplanter are presented in this chapter.

4.1 Evaluation of seedling ejection by contemplated methods

Two different methods of seedling ejections were tried to explore their feasibility
for design of the automatic vegetable transplanter.

4.1.1 Seedling ejection from protray by gravity

The seedlings were grown in open bottom type protray with growing medium of
coir pith + vermicompost (4:1) (media 1), coir pith + vermicompost (4:1) + paper cone
(media 2), soil + vermicompost (4:1) (media 3), and soil + vermicompost (4:1) + paper
cone (media 4), and tested on an exclusively designed test stand as explained in section
3.2.1. The agronomical parameters like plant height, stem diameter, number of leaves and
root weight of nursery grown in open bottom type protray and conventional protray were
measured to see whether there is any variation, and it was found that no significant
difference between them before the ejection tests were initiated. The results of the gravity
ejections are shown in figure 4.1. It was observed that the seedlings grown in soil +
vermicompost (4:1) and in paper cone fell down successfully due to the weight of the
seedling being high when compared to other media and since the paper cone prevents
media from sticking to the protray surface. However, as explained in section 3.2, this
method was found not practicable in the contemplated machine.

4.1.2 Seedling ejection by clawing out from the top side of the protray cell

A four bar mechanism was designed to extract the seedling from the tray cell and
release it to fall free to the ground. But, while the seedling was taken out from the cells

119
the root ball disintegration was beyond feasible limits and hence no further testing than
the preliminary trials were done.

Fig 4.1 Effect of gravity falling with open bottom type protray

4.2 Logit modeling of seedling ejection performance of motorized plunger mechanism

4.2.1 Logit modeling of binary performance

In ordinary regression,  = E(Y) is a linear function of x, i.e, y =  + x. For a

binary response an analogous model is (x) =  + x. This is called a linear probability
model, because the probability of success changes linearly in x. The parameter ‘’
represents the change in the probability per unit change in x. This model is a Generalized
Linear Model (GLM) with binomial random component and identity link function. This
model is simple, but probabilities fall between 0 and 1, whereas linear functions take
values over the entire range. It hence predicts (x)  0 and (x)  1 for sufficiently large
or small x values. Relationships between (x) and x are usually nonlinear rather than
linear. A fixed change in x may have less impact when  is near 0 or 1 than when  is
near the middle of its range. In practice, (x) often either increases continuously or
decreases continuously as x increases. An S shaped curve is often a realistic shape of this
relationship between (x) and x. The mathematical function with this shape has
120
 αβx  π(x) 
e
formula π(x)  which when written in another form, log   α  βx is

αβx  1  π(x) 
e
called the logistic regression model. This is a special case of a GLM and the random
component for the outcomes (success or failure) has a binomial distribution, which in this
 π 
study is success or failure of an event. The link function is the logit function log 
 (1 π) 

of , whereas  is restricted to the 0 to 1 range that the logit can be any real number. The
real numbers are also the potential range for linear predictions (such as  + x). The
parameter  determines the rate of increase or decrease of the curve. An experimental
design with binary response can hence analyze the influence of factors similar to other
regressive models.

4.2.2 Logit modeling of seedling ejection

The experiments on motorized plunger in popping out the plugs to optimize the
variables such as age of seedling, effect of medium, plunger diameter and plunger speed
was explained in section 3.4.3. The recorded raw data was initially brought into memory
and the proportion of successful ejection to the total number of ejection was found. This
was stored as probability of successful ejection. Next, the success and failure of ejection
was arranged as a matrix with rows as the treatment combinations. The function “glm”
standing for generalized linear model was invoked to model this response matrix against
the full interactive combinations of the four influencing factors. The built model was
summarized as shown in the model listing - I. In an attempt to simplify the model, the
built model, was updated to have only additive effect of four factors. These two models
were compared and it was found that the simpler additive model was sufficiently
explaining the binomial response. Hence, the linear model was stepped through by
systematic removal of factors based on AIC (Akaike Information Criteria) value and the
minimal adequate model was found out using the “step” function of the ‘R’ statistical
tool. The stepwise logistic regression was continued and finally it was found that the
observed binomial response of ejecting was adequately modelled by an additive model of
medium, age and plunger diameter. The process of stepwise regression was checked for

121
adequacy in each step. Having the model in hand, the fitted/predicted response was
calculated. These were the probabilities of successful ejection. Using conventional
plotting technique, the observed as well as predicted probability of successful ejection
(Fig.4.2) was plotted against the medium. The predicted values are shown as lines and
observed probability as points.

In figure 4.2, the predicted values are perfectly fitting the observed probability
values and for the 6 mm diameter plunger, the probability of successful ejection was
more than 90 per cent for all the media. While using 5 and 4 mm plunger diameter for
ejection, the probability of success is 80 to 85 per cent in all the media. But for 3 mm
plunger diameter, the observed value is less than predicted value for M1 (coir pith) and
M2 (coir pith + vermicompost 4:1). The plunger instead of ejecting, it penetrates the
medium and therefore the probability of ejection is less than 80 per cent. However for
media that have soil or sand, the probability of ejection is more because the soil mixture
provides solidity to the medium and the probability of ejection is more than 80 per cent.
Generally, the diameter of plunger increases the probability of successful seedling
ejection, because the bigger diameter plunger reduces penetration into the media.

Figure 4.3 shows the same model predictions for the ejection performance as
against the age of seedling. It may be seen from the figure that the probability of successful
ejection increased with plunger diameter and decreased with age of seedling. The probability
of successful ejection of 20 days old seedling is found to be 80 to 90 per cent where as that
of 30 and 40 days old seedling was 65 to 80 and 40 to 60 per cent respectively. Larger
plants of more age tend to entangle with each other preventing proper ejection. Smaller
plunger diameters penetrated the plug rather than ejecting them and hence have brought
about lesser probability of success.

122
 1.0
Probability of successful drops
0.6 0.8
6 mm dia
5 mm dia
4 mm dia
0.4 0.2
0.0 3 mm dia

M1 M2 M3 M4 M5
Growth medium

Fig 4.2 Effect of medium on plant ejection

1.0
Probability of successful drops
0.4 0.6 0.8

6 mm dia
5 mm dia
4 mm dia
0.2

3 mm dia
0.0

20 25 30 35 40
Age of seedling, days

Fig 4.3 Effect of age on plant ejection

123
 Model listing - I

Logit modeling of experimental seedling ejection performance

> #Logistic regression - ejecting performance

> ejectcounts <-

read.table("e:/download/SivakumarProject/pluglogit.csv",sep=',',header=T)

> attach(ejectcounts)

The following object is masked from uprgtcounts:

medium, success

> p<-success/25 #proportion of successful drops to total)

> y<-cbind(success, 25-success) # Bind the proportions for using in glm binomial
modeling

> #Do a binomial logit full model

> model<-glm(y~age*medium*pspeed*pdia,binomial)

> #Update the model without any interactions

> model1<-glm(y~age+medium+pspeed+pdia,binomial)

> summary(model1)

Call:

glm(formula = y ~ age + medium + pspeed + pdia, family = binomial)

Deviance Residuals:

Min 1Q Median 3Q Max

-1.5507 -0.3231 -0.0005 0.3962 1.4614

Coefficients:

Estimate Std. Error z value Pr(>|z|)

(Intercept) 2.001e+00 7.784e-02 25.707 < 2e-16 ***

124
ageA2 -8.330e-01 5.901e-02 -14.115 < 2e-16 ***

ageA3 -1.614e+00 5.635e-02 -28.653 < 2e-16 ***

mediumM2 -4.606e-02 6.196e-02 -0.743 0.4572

mediumM3 5.842e-01 6.649e-02 8.786 < 2e-16 ***

mediumM4 5.474e-01 6.614e-02 8.277 < 2e-16 ***

mediumM5 5.256e-01 6.593e-02 7.972 1.56e-15 ***

pspeedS2 -8.718e-02 5.141e-02 -1.696 0.0899

pspeedS3 -4.663e-02 5.163e-02 -0.903 0.3664

pdiaD2 -8.119e-17 6.206e-02 0.000 1.0000

pdiaD3 -3.015e-01 6.009e-02 -5.018 5.23e-07 ***

pdiaD4 -5.025e-01 5.913e-02 -8.497 < 2e-16 ***

---

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

Null deviance: 1377.97 on 539 degrees of freedom

Residual deviance: 148.91 on 528 degrees of freedom

AIC: 1905.6

Number of Fisher Scoring iterations: 4

> #Comparing the two models; model wo any interaction is sufficient

> anova(model1,model, test="Chi")

Analysis of Deviance Table

Model 1: y ~ age + medium + pspeed + pdia

Model 2: y ~ age * medium * pspeed * pdia

Resid. Df Resid. Dev Df Deviance Pr(>Chi)

125
 1 528 148.914

2 360 71.246 168 77.668 1

> #Stepping thru to find the optimal combination

> step(model1)

Start: AIC=1905.58

y ~ age + medium + pspeed + pdia

Df Deviance AIC

- pspeed 2 151.79 1904.5

<none> 148.91 1905.6

- pdia 3 253.96 2004.6

- medium 4 336.34 2085.0

- age 2 1104.93 2857.6

Step: AIC=1904.46

y ~ age + medium + pdia

Df Deviance AIC

<none> 151.79 1904.5

- pdia 3 256.81 2003.5

- medium 4 339.17 2083.8

- age 2 1107.60 2856.3

Call: glm(formula = y ~ age + medium + pdia, family = binomial)

Coefficients:

(Intercept) ageA2 ageA3 mediumM2

1.956e+00 -8.328e-01 -1.614e+00 -4.605e-02

mediumM3 mediumM4 mediumM5 pdiaD2

126
 5.841e-01 5.472e-01 5.255e-01 -8.897e-18

pdiaD3 pdiaD4 -3.014e-01 -5.023e-01

Degrees of Freedom: 539 Total (i.e. Null); 530 Residual

Null Deviance: 1378

Residual Deviance: 151.8 AIC: 1904

> #Found the optimal model as additive with pspeed eliminated

> #Final model has only age+medium+pdia

> model2<-update(model1, ~.-pspeed)

> summary(model2)

Call:

glm(formula = y ~ age + medium + pdia, family = binomial)

Deviance Residuals:

Min 1Q Median 3Q Max

-1.57453 -0.33109 0.01389 0.38377 1.45914

Coefficients:

Estimate Std. Error z value Pr(>|z|)

(Intercept) 1.956e+00 7.169e-02 27.283 < 2e-16 ***

ageA2 -8.328e-01 5.901e-02 -14.114 < 2e-16 ***

ageA3 -1.614e+00 5.634e-02 -28.650 < 2e-16 ***

mediumM2 -4.605e-02 6.195e-02 -0.743 0.457

mediumM3 5.841e-01 6.648e-02 8.785 < 2e-16 ***

mediumM4 5.472e-01 6.613e-02 8.276 < 2e-16 ***

mediumM5 5.255e-01 6.592e-02 7.971 1.57e-15 ***

pdiaD2 -8.897e-18 6.205e-02 0.000 1.000

127
pdiaD3 -3.014e-01 6.009e-02 -5.017 5.25e-07 ***

pdiaD4 -5.023e-01 5.912e-02 -8.496 < 2e-16 ***

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

(Dispersion parameter for binomial family taken to be 1)

Null deviance: 1377.97 on 539 degrees of freedom

Residual deviance: 151.79 on 530 degrees of freedom

AIC: 1904.5

Number of Fisher Scoring iterations: 4

> #Predict with "response" wil give the predicted proprtions not the fitted logits

> q=predict(model2,type="response") #giving the predicted "respons ..." ...

> predprobs =as.vector(q)

> windows()

> #par(mfrow=c(1,2))

> #plot(c("M1","M2","M3","M4","M5"),predprobs[c(1,37,73,109,145)], type="l",xlab =

"Growth Medium", ylab = "Probability of succes ..." ...

> axis(1, at=c(1:5),labels=c("M1","M2","M3","M4","M5"))

> lines(c(1:5),predprobs[c(4,40,76,112,148)], type="l", ylim=c(0,1), lwd=2, col=3)

> lines(c(1:5),predprobs[c(7,43,79,115,151)], type="l", ylim=c(0,1), lwd=2, col=4)

> lines(c(1:5),predprobs[c(10,46,82,118,154)], type="l", ylim=c(0,1), lwd=2, col=5)

> calcprobs=success/25

> points(c(1:5), calcprobs[c(1,37,73,109,145)], pch=18, col=2)

> points(c(1:5), calcprobs[c(4,40,76,112,148)], pch=15, col=3)

> points(c(1:5), calcprobs[c(7,43,79,115,151)], pch=17, col=4)

> points(c(1:5), calcprobs[c(10,46,82,118,154)], pch=16, col=5)

128
> #Drawing the legends for each variety at different locations

> legend(4, 0.6, c("6 mm dia","5 mm dia", "4 mm dia", "3 mm dia"), lty =1,
pch=c(18,15, ....

> #Open another graph window

> windows()

> #par(ask=TRUE)

> plot(c(20,30,40),predprobs[c(1,181,361)], type="l",xlab = "Age of seedling, DAS",

ylab = "Probability of successful drops", ylim=c ....

> axis(1, at=c(20,30,40),labels=c(20,30,40))

> lines(c(20,30,40),predprobs[c(4,184,364)], type="l", ylim=c(0,1), lwd=2, col=3)

> lines(c(20,30,40),predprobs[c(7,187,367)], type="l", ylim=c(0,1), lwd=2, col=4)

> lines(c(20,30,40),predprobs[c(10,190,370)], type="l", ylim=c(0,1), lwd=2, col=5)

> points(c(20,30,40), calcprobs[c(1,181,361)], pch=18, col=2)

> points(c(20,30,40), calcprobs[c(4,184,364)], pch=15, col=3)

> points(c(20,30,40), calcprobs[c(7,187,367)], pch=17, col=4)

> points(c(20,30,40), calcprobs[c(10,190,370)], pch=16, col=5)

> legend(35, 0.4, c("6 mm dia","5 mm dia", "4 mm dia", "3 mm dia"), lty =1,

pch=c(18,15,17,16),col=c(2,3,4,5), bty="n")

> #lines(c(15,20,25),predprobs[13:15], type="l", ylim=c(0,1), lty= 2, lwd=2)

> #lines(c(15,20,25),predprobs[16:18], type="l", ylim=c(0,1), lty= 3, lwd ....

129
4.3 Effect of growth medium and age on agronomical parameters

A clear understanding of agronomic practices of vegetable seedling growing in

protrays and other conventional practices in terms of growth medium, methods of nursery
raising techniques and age of seedling opt for transplanting in the field are required for
the design of automatic vegetable transplanter.

In section 3.5, the experiments on agronomical parameters of the seedlings grown in

protrays were explained. Five different combinations of the growth media were used to grow
tomato seedling in the protrays. Biometric observations at 20, 30 and 40 days of plant growth
as influenced by the five growth media were recorded as influencing plant height, stem
thickness, number of leaves, plug weight, stem weight, root weight and pulling force.

4.3.1 Effect on plant height

Table 4.1 presents the ANOVA on seedling height as influenced by the different
media and age. The selected variables significantly influenced the seedling height. This is
due to variation of naturally available nutrients in the selected growth media. 40 days old
seedlings (A3) as grown in coir pith + vermicompost 4:1 (M2) obtained the maximum
height of 230 mm. The 20 days old seedling (A2) with coir pith + soil 1:1 (M3) grew
only to a height of 110 mm (Fig.4.4). Interactions of the variables also were proved to be
significant. This was due to the fact that M3, M4 and M5 mediums were not quite
responsive to age as that of M1 and M2. Kumar and Raheman, 2011; Satpathy and Garg,
2008; and Brewer, 1995 confirmed these results.

Table 4.1 Analysis of variance on seedling height

Source df SS MS F PROB
Total 44 40189.514031 913.398046 10.3016
Treatment 14 37529.541098 2680.681507 30.2336 0.000 **
Age (A) 2 20450.532111 10225.266056 115.3237 0.000 **
Medium (M) 4 12949.075476 3237.268869 36.5109 0.000 **
AxM 8 4129.933511 516.241689 5.8223 0.011 **
Error 30 2659.972933 88.665764 1.0000
(CV = 0.85%, ** - significant at 1 % level)
130
4.3.2 Effect on stem thickness

The stem thickness was also significantly influenced by the medium and age (Table
4.2). Figure 4.5 depicts that the stem thickness was quite high in 40 days old seedlings
(A3) with coir pith + vermicompost 4:1 (M2) having the maximum thickness of 3.6 mm.
The minimum stem thickness (2.4 mm) was found in 20 days old seedling (A2) with coir
pith alone (M1). The results were almost similar to that on plant height as influenced by
age and medium. The interactions were also significant explaining that the response was
not uniform in all media as influenced by the age. Similar results were reported by Singh
et al., 2007. The increase of stem thickness and plant height with age needs no
explanation, however is shown to be statistically significant.

Table 4.2 Analysis of variance on stem thickness

Source df SS MS F PROB

Total 44 4.983444 0.113260 3.7672

Treatment 14 4.081511 0.291537 9.6971 0.000 **

Age (A) 2 1.773924 0.886962 29.5020 0.000 **

Medium (M) 4 1.080889 0.270222 8.9881 0.005 **

AxM 8 1.226698 0.153337 5.1003 0.017 *

Error 30 0.901933 0.030064 1.0000

(CV = 0.55% ** - significant at 1 % level)

4.3.3 Effect on number of leaves

The number of leaves (Table 4.3) was significantly influenced by the medium and
age. The number of leaves was found to be high in 40 days old seedlings (A3) grown in
coir pith + vermicompost 4:1 (M2) and the minimum number of leaves were found in 20
days old seedling (A1) grown in coir pith alone (M1& M3) and 30 days old seedling (A3)
with soil mixtures (M4 & M5) (Singh et al., 2007). The interactions were significant but
not explainable clearly (Fig.4.6).

131
Fig 4.4 Effect of different media on seedling height

Fig 4.5 Effect of different media on stem thickness

132
Table 4.3 Analysis of variance on number of leaves

Source df SS MS F PROB

Total 44 11.407880 0.259270 4.4069

Treatment 14 9.642880 0.688777 11.7073 0.000 **

Age (A) 2 3.114880 1.557440 26.4721 0.000 **

Medium (M) 4 4.660480 1.165120 19.8037 0.000 **

AxM 8 1.867520 0.233440 3.9678 0.034 *

Error 30 1.765000 0.058833 1.0000

CV = 0.65% ** - significant at 1 % level

Altogether, the difference in seedling height, stem thickness and number of leaves
is owing to nutrient availability of the growth media. The vermicompost provides
additional nutrient to the growth medium (M2), hence the coir pith + vermicompost 4:1
growth medium was found to best for growth. The seedling height, stem thickness and
number of leaves may be considered as indicators of the health of the seedling. This may
have an effect on reestablishment of plant growth in the field after transplanting.

Fig 4.6 Effect of different media on number of leaves

133
4.3.4 Effect on plug weight
The experiment on optimization of growth medium and age for better growing of
seedling was explained in section 3.5. A linear model was attempted on the plug weight
as influenced by the two factors namely the type of growth medium and age. A diagnostic
was attempted on the developed model (Fig.4.7). On analysing the raw data of plug
weight for normalcy, it was found that the distribution was not normal. But normalcy of
data is mandatory for building a linear model. The Q - Q plot (Fig.4.7) shows that the
residuals were not normally distributed.

A box - cox analysis was hence attempted on the model to find an appropriate
transformation coefficient (). Figure 4.8 illustrates the log likely hood for different
 values for the full model being considered. The transformatic coefficient was found to
be -1.23, implying that the transformation necessitated on the plug weight is the power of
- 1.23. Using the above transformation for the plug weight response, a full model
involving the growth medium and age was tried next.

Figure 4.9 illustrates the diagnostic plot of the new model after transformation,
wherein no evidence was found on any trend in the distribution of residual against the
fitted value. The Q - Q plot also proved that the response had a perfect normal
distribution, which was quite anticipated because of the box - cox transformation.

Table 4.4 shows the ANOVA on plug weight as influenced by growth medium and
age of seedling. The age was found to be non significant on the plug weight. This proved
that the age does not significantly change the root mass as anticipated. Now a simple
model was built using only medium as factor on the plug weight.

Table 4.4 Analysis of variance on plug weight

Source Df Sum Sq Mean Sq F value Pr (>F)
Medium (M) 4 353.1 88.28 58.02 2e-16 ***
Age (A) 2 3.2 1.60 1.05 0.356
AxM 8 196.8 24.60 16.17 1.94e-12 ***
Residuals 60 91.3 1.52
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
134
 Residuals vs Fitted Scale-Location
48

2.0
48
4

S tandardized res iduals

66

1.5
68
68
2
Residuals

1.0
0

0.5
-2

66

0.0
10 12 14 16 18 10 12 14 16 18

Fitted values Fitted values

Constant Leverage:
Normal Q-Q Residuals vs Factor Levels

48 48
4
Standardized residuals

Standardized residuals

68 68
2
2

0
0

-2
-2

66
66
-4

Medium :
-2 -1 0 1 2 M1 M2 M3 M4 M5

Theoretical Quantiles Factor Level Combinations

Fig 4.7 Diagnostics on model of plug weight (first model tried)

135
 95%

55.5
55
95%

55.0
Inset showing the exact value

log-Likelihood
50

54.5
45

54.0
-1.30 -1.28 -1.26 -1.24 -1.22 -1.20
log-Likelihood

40

Box cox factor

35
30
25
20

-2 -1 0 1 2
Box cox factor

Fig 4.8 Determination of Box - Cox transformation parameter (plug weight)

Figure 4.10 shows the Tukey plots relevant to the influence of growth media on the
plug weight of the seedling. Whichever range of mean differences between pairs cross
zero are on par. It may be noted that, M1 (coir pith) and M2 ( coir pith + vermicompost
at 4:1); M5 (soil + sand + coir pith at 2:1:1) and M3 (coir pith + soil 1:1) ; M4 (clay soil
+ sand + coir pith 1:1:1) and M3 (coir pith + soil 1:1); and finally M4 (clay soil + sand + coir
pith 1:1:1) and M5 (soil + sand + coir pith 2:1:1) were on par influencing the root weight.
Obviously, the media with soil and sand are heavier than those without soil and sand.

136
 Residuals vs Fitted Scale-Location
48
4

2.0
66

S tandardized res iduals

66
2

1.5
68
Residuals

1.0
-2

68

0.5
-4

48

0.0
-6

-22 -20 -18 -16 -14 -12 -22 -20 -18 -16 -14 -12

Fitted values Fitted values

Constant Leverage:
Normal Q - Q Residuals vs Factor Levels
4

66 66
2
2
Standardized residuals

Standardized residuals

0
0

-2
-2

68
68
-4
-4

48
48

Medium :
-2 -1 0 1 2 M5 M4 M3 M2 M1

Theoretical Quantiles Factor Level Combinations

Fig 4.9 Diagnostics on model of plug weight (model appropriate)

137
M1-M2

M3-M2

M5-M2

M4-M2

M3-M1
M1 - Coir pith

M5-M1 M2 - Coir pith +

vermicompost
(4:1)
M4-M1
M3 - Coir pith +
soil (1:1)
M5-M3
M4 - Clay soil +
sand + coir pith
M4-M3 (1:1:1)

M5 - Soil + sand
M4-M5 + coir pith (2:1:1)

0 2 4 6

Differences in mean levels of medium, g

Fig 4.10 Tukey mean comparison on the root weight of nursery plants as influenced
by growth medium

It is evident from figure 4.11 that the plug weight was found to be more in M4
(coir pith + sand + clay soil) 17.86 g and the lowest at 10.68 g for M1 (coir pith) for the
20 days old seedling. The growth medium exhibited significant difference in plug weight
due to its own weight of the growth medium, since the soil, coarse sand and clay soil
compositions having more weight than coir pith and coir pith + vermicompost
(Burgoyne, 1987; and Brewer, 1995).

138
 Model listing - I is the sample script of the statistical routine, which is self
explanatory of the above explained process of analysis. The results show that using aged
seedlings will not help in better transplanting dynamics as influenced by plug weight.

Fig 4.11 Effect of plug weight on different medium and age

139
 Model listing - II

Plug weight analysis on growth medium and age

> #Reading the data in

> dat1= read.table("e:/download/SivakumarProject/pullingforce.csv", sep=',', header =

TRUE)

> #Plugweight analysis

> #Trying the full anova model

> a1=aov(plugwt~medium*age, data = dat1)

> summary(a1)

Df Sum Sq Mean Sq F value Pr(>F)

medium 4 233.41 58.35 58.02 < 2e-16 ***

age 2 2.11 1.06 1.05 0.356

medium:age 8 130.08 16.26 16.17 1.94e-12 ***

Residuals 60 60.35 1.01

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> #Open another graphical window and split screen into 2 by 2 areas

> windows()

> layout(matrix(c(1,2,3,4),2,2))

> #Diagnostics plot

> plot(a1)

> #Normalcy assumption is not obeyed

> #Going for Box cox transformation test

> library(MASS)

> windows()

> #Drawing box cox diagnostics

140
> boxcox(a1)

> windows()

> #Drawing box cox with more resolution

> boxcox(a1,lambda = seq(-1.3, -1.2, 0.01))

> #Applying the lamda transformation to response variable

> #modeling in full

> a1=aov(plugwt*(-1.23)~medium*age, data = dat1)

> summary(a1)

Df Sum Sq Mean Sq F value Pr(>F)

medium 4 353.1 88.28 58.02 < 2e-16 ***

age 2 3.2 1.60 1.05 0.356

medium:age 8 196.8 24.60 16.17 1.94e-12 ***

Residuals 60 91.3 1.52

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> windows()

> layout(matrix(c(1,2,3,4),2,2))

> #Diagnostics plot again

> plot(a1)

> #Now diagnostics better

> #Building a simpler model excluding non significant age

> a1=aov(plugwt*(-1.23)~medium, data = dat1)

> summary(a1)

Df Sum Sq Mean Sq F value Pr(>F)

medium 4 353.1 88.28 21.21 1.72e-11 ***

Residuals 70 291.3 4.16

141
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> #Open yet another graphical window and split screen into 2 by 2 areas

> windows ()

> #layout (matrix(c(1,2,3,4),2,2))

> #Plot the tukey test plots of medium

> #whichever range of mean differences between pairs cross zero are on par

4.3.5 Effect on Stem weight

The linear model was built and analysed by way of an ANOVA, the model was
checked for adequacy in terms of the residuals plot as well as the Q - Q plot to check the
normalcy of the observed data. Figure 4.12 shows the diagnostic plot of the model built
for analysing the stem weight as influenced by the growth medium and age. The Q - Q
plot showed that the residuals were not normally distributed. Hence here also the same
procedure described in the previous section 4.3.4 was followed. The box- cox analysis
was attempted on the model to find an appropriate transformation coefficient () and
again the diagnostic plot was developed. The figure 4.13 shows the log likely hood for
different  values for the full model being considered. The transformatic coefficient was
found to be - 0.75, implying that the transformation necessitated on the plug weight is the
power of - 0.75. Using the above transformation for the plug weight response, a full
model involving the growth medium and age was developed.

Figure 4.14 shows that the diagnostic plot was found to be better and the residuals
are normally distributed. The Q - Q plot showed good linearity of data proving the
adequacy of the model. Similarly the residual plots did not show any definite trend and
were truly random showing that the model built was adequate. Nextly a simpler model
was developed excluding insignificant age factor.

Table 4.5 summaries the ANOVA on stem weight as influenced by growth

medium. The age was found to be non significant of the stem weight. It was inferred from
the ANOVA (Table 4.5) that the growth medium as well as the interaction effect
significantly influenced the stem weight and age is non significant.
142
Table 4.5 Analysis of variance on stem weight

Source Df Sum Sq Mean Sq F value Pr (>F)

Medium (M) 4 7.976 1.9939 41.091 < 2e-16 ***
Age (A) 2 1.656 0.8278 17.059 0.048
AxM 8 3.031 0.3789 7.808 4.02e-07 ***
Residuals 60 2.911 0.0485
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Fig 4.12 Diagnostics on model of stem weight (first model tried)

143
 95%
Inset showing the exact value

-10

-4.5
log-Likelihood

-5.0
-20

-5.5
-30
log-Likelihood

-6.0
95%
-40

-0.80 -0.78 -0.76 -0.74 -0.72 -0.70

-50
-60

-2 -1 0 1 2

Fig 4.13 Determination of Box - Cox transformation parameter (stem weight)

The Tukey plot in figure 4.15 indicated the mean comparison by way of Tukey
test. The different growth media were showing significant, and hence the means were
statistically on par in affecting the response, here the stem weight. The sample plot shows
the deviations between the levels of growth medium contain zero and indicate on par
status. M3 (coir pith + soil 1:1) - M1 (coir pith), M5 (soil + sand + coir pith 2:1:1) - M3
(coir pith + soil 1:1) , M4 (clay soil + sand + coir pith 1:1:1) - M3 (coir pith + soil 1:1)
and M4 (clay soil + sand + coir pith 1:1:1) - M5 (soil + sand + coir pith 2:1:1) were on
par status.

Figure 4.16 showed that the growth medium only influences the stem weight due to
nutrient availability. The highest stem weight (2.7 g) was found in M2 (coir pith +
vermicompost 4:1) and lowest was M3 (coir pith + soil 1:1), M4 (clay soil + sand + coir
pith 1:1:1) and M5 (soil + sand + coir pith 2:1:1) (1 g) on 20 days old seedling. This

144
 weight differences were due to vermicompost providing additional nutrient to the
seedling (Burgoyne, 1987; and Brewer, 1995).

Residuals vs Fitted Scale-Location

2.0
54
0.4

Standardized res iduals

28
29

1.5
Residuals

0.0

1.0
-0.4

0.5
29
28
-0.8

54

0.0
-2.5 -2.0 -1.5 -1.0 -2.5 -2.0 -1.5 -1.0

Fitted values Fitted values

Constant Leverage:
Normal Q -Q Residuals vs Factor Levels
3

3
2

2
Standardized residuals

Standardized residuals

1
1

0
0

- 4 -3 -2 -1
- 4 -3 -2 -1

29
29 28
28
54
54

Medium :
-2 -1 0 1 2 M2 M1 M4 M3 M5

Theoretical Quantiles Factor Level Combinations

Fig 4.14 Diagnostics on model of stem weight (model appropriate)

145
 M1-M2

M3-M2

M5-M2

M4-M2
M1 - Coir pith
M3-M1
M2 - Coir pith +
vermicompost
M5-M1 (4:1)

M4-M1 M3 - Coir pith +

soil (1:1)
M5-M3 M4 - Clay soil +
sand + coir pith
M4-M3 (1:1:1)

M5 - Soil + sand +
M4-M5 coir pith (2:1:1)

0.0 0.5 1.0

Differences in mean levels of medium, g

Fig 4.15 Tukey mean comparison on the stem weight of seedling as influenced by
growth medium and age

Fig 4.16 Effect of stem weight on different medium and age

146
4.3.6 Effect of Root weight

Here also the linear model was built on the root weight. Figure 4.17 showed the
diagnostic model first tried, it was found that from the Q - Q plot the residuals were not
normally distributed as similar to the previous section in 4.3.4. Therefore the box - cox
analysis was tried on the model to find an appropriate transformation coefficient ().
Figure 4.18 illustrates the log likely hood for different  values for the full model being
considered. The transformation coefficient was found to be 0.85. Using the above
transformation for the root weight response, a full model involving the growth media and
age was built. The best model was put under diagnostics to find out whether the built
linear model was adequate. Figure 4.20 shows the diagnostic model (improved) and the Q
- Q plot also proved that the response had a perfect normal distribution and the model
was improved due to the box - cox transformation. Here also the insignificant variable is
removed and a model built.

Table 4.6 shows the ANOVA on root weight as influenced by medium and the age
of seedling was found to be insignificantly influencing.

Table 4.6 Analysis of variance on root weight

Source Df Sum Sq Mean Sq F value Pr (>F)

Medium (M) 4 2.4376 0.6094 26.021 1.56e-12 ***

Age (A) 2 0.0683 0.0342 1.459 0.24062

AxM 8 0.7152 0.0894 3.817 0.00111 **

Residuals 60 1.4052 0.0234

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Figure 4.19 shows the Tukey plots relevant to the influencing factor on root weight.
M1 (coir pith) - M2 ( coir pith + vermicompost 4 :1), M5 (soil + sand + coir pith 2:1:1)
- M3 (coir pith + soil 1:1) , M4 (clay soil + sand + coir pith 1:1:1) - M3 (coir pith + soil
1:1) and M4 (clay soil + sand + coir pith 1:1:1) - M5 (soil + sand + coir pith 2:1:1) were

147
on par status. The growth medium reveals significant difference in root weight due to
nutrient availability of the growth medium.

The root weight (0.5 g) (Fig.4.21) was found in M2 (coir pith + vermicompost
4:1) on 30 days old seedling and lowest was M3 (coir pith + soil 1:1), M4 (clay soil +
sand + coir pith 1:1:1) and M5 (soil + sand + coir pith 2:1:1) (1 g) on 20 days old
seedling. This weight differences are because of vermicompost compositions providing
additional nutrient to the seedling.

Residuals vs Fitted Scale-Location

0.15

52

1.5
52 65

Standardized res iduals

37
37
0.05

1.0
Residuals

-0.05

0.5
65
-0.15

0.0

0.30 0.35 0.40 0.45 0.50 0.55 0.30 0.35 0.40 0.45 0.50 0.55

Fitted values Fitted values

Constant Leverage:
Normal Q - Q Residuals vs Factor Levels
3
3

52 52
Standardized residuals

Standardized residuals

2
2

37 37
1
1

0
0

-1
-1

-2
-2

65
65

Medium :
-2 -1 0 1 2 M4 M5 M1 M3 M2

Theoretical Quantiles Factor Level Combinations

Fig 4.17 Diagnostics on model of root weight (first model tried)

148
 -10.0
-10
95%

-10.5
-15

log-Likelihood

-11.0
-20

-11.5
log-Likelihood

95%
-12.0
-25

0.80 0.85 0.90 0.95 1.00

-30
-35
-40

-2 -1 0 1 2

Fig 4.18 Determination of Box - Cox transformation parameter (root weight)

M1-M2

M3-M2

M5-M2

M4-M2

M3-M1
M1 - Coir pith

M5-M1 M2 - Coir pith +

vermicompost (4:1)
M4-M1
M3 - Coir pith + soil
M5-M3 (1:1)

M4 - Clay soil + sand

M4-M3 + coir pith (1:1:1)

M4-M5 M5 - Soil + sand +

coir pith (2:1:1)
-0.1 0.0 0.1 0.2 0.3
Differences in mean levels of medium, g

Fig 4.19 Tukey mean comparison on the root weight of nursery plants as influenced
by growth medium
149
 Residuals vs Fitted Scale-Location
38

1.5
37 65
37

Standardized res iduals

0.2
Residuals

1.0
0.0
-0.2

0.5
65
-0.4

38

0.0
-1.2 -1.0 -0.8 -0.6 -1.2 -1.0 -0.8 -0.6

Fitted values Fitted values

Constant Leverage:
Normal Q-Q Residuals vs Factor Levels
37 37
2
Standardized residuals

Standardized residuals
2

1
1

0
0

-1
-1

-2
-2

65
65
-3

38
38
-3

Medium :
-2 -1 0 1 2 M4 M5 M1 M3 M2

Theoretical Quantiles Factor Level Combinations

Fig 4.20 Diagnostics on model of root weight (model appropriate)

150
 Fig 4.21 Effect of plug weight on different medium and age

4.4 Effect of pulling force on growth medium and age

The pulling force was measured using a spring balance as explained in the section
3.6. The pulling force affected the successful seedling ejection. If the ejection force is
excessive the plunger may pierce rather than eject. The two important factors that may
influence the pulling force are growth media and age. The full ANOVA of this model
was tried. The full model was built and the general diagnostic procedures were done by
plotting the normal Q - Q plot, the residuals and the leverages (Fig.4.22). As may be
seen, the residuals are randomly associated with the fitted values showing that there is no
bias or trend in the observations of the experiment conducted. The normalcy of the
observations was proven in the Q - Q plot with the points lying on the straight line and
hence proven to be normally distributed. Being so, the diagnostic plot proved the full
linear model that is fitted, is found adequate. The insignificant variable is removed and a
model built as explained in the previous section 4.3.4.

Table 4.7 shows the ANOVA on pulling force as influenced by both factors. The
age of seedling was also found to be significantly influencing. The interactions of two
variables are also significant. The age of seedling could have influenced the roots on the
cell wall periphery causing a larger force.
151
 Residuals vs Fitted Scale-Location

S t a n d a r d iz e d r e s id u a l s
1

1.5
52
10

9
5

1.0
Residuals

0.5
-10

9 52
1

0.0
-20

110 120 130 140 150 160 110 120 130 140 150 160

Fitted values Fitted values

Constant Leverage:
Normal Q - Q Residuals vs Factor Levels
2

2
1
1
Standardized residuals

Standardized residuals

0
0

-1
-1

-2

9
-2

529 52
1
-3

1
-3

Medium :
-2 -1 0 1 2 M2 M1 M3 M5 M4

Theoretical Quantiles Factor Level Combinations

Fig 4.22 Diagnostics on model of pulling force (model appropriate)

152
Table 4.7 Analysis of variance on pulling force

Source Df Sum Sq Mean Sq F value Pr (>F)

Medium (M) 4 21298 5324 121.193 < 2e-16 ***

Age (A) 2 395 97 4.495 0.0152 *

AxM 8 1864 233 5.304 4.8e-05 ***

Residuals 60 1864 233 5.304 4.8e-05 ***

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Figure 4.23 shows that the Tukey plot of the mean comparison of the factors
considered. Only M5 (soil + sand + coir pith 2:1:1) - M3 (coir pith + soil 1:1) were on
par, with all the other treatments different from each other. This result is very much
different from the results on the other agronomical factors. The root configuration in each
medium and age was proven to be different causing difference in force of extraction.

M1-M2

M3-M2

M5-M2

M4-M2
M1 - Coir pith
M3-M1 M2 - Coir pith +
vermicompost
M5-M1 (4:1)

M3 - Coir pith +
M4-M1 soil (1:1)

M5-M3 M4 - Clay soil +

sand + coir pith
(1:1:1)
M4-M3
M5 - Soil + sand
M4-M5 + coir pith (2:1:1)

0 10 20 30 40 50
Differences in mean levels of medium, g
Fig 4.23 Tukey mean comparison on the pulling force of nursery plants as
influenced by growth medium and age
153
 Figure 4.24 shows the effect of pulling force and it may be noted that the soil
mixture compositions have more pulling force (1.6 N) than coir pith + vermicompost
(1.03N). This is due to soil mixtures having stickiness which may stick on to the protray.
Moreover, the plug weight was also more in soil mixture mediums compared to coir pith
and coir pith + vermicompost mediums (Burgoyne, 1987; and Osamu Sakaue, 1992).

4.5 Modeling of root ball weight loss as influenced by medium, height of fall,
hardening and cushioning.

In section 3.7.2, the drop test for weight loss was explained wherein the loss of
weight was modelled against the day of plug hardening, the thickness of cushion used to
absorb the falling shock and the height of drop. A linear model was tried on root ball
weight loss as influenced by the above said factors. A diagnostic was attempted on the
developed model (Fig.4.25.) The Q - Q plot shows that the residuals were not normally
distributed. A box - cox analysis was attempted on the developed model to verify an
appropriate transformation coefficient lamda (). Applying the lamda transformation to
response variables, the full model was tried and diagnostic plot drawn again. It was found
to be still bad and hence a square root transformation was effected and diagnostic plot
once again drawn. It was now found that the diagnostics was better (Fig.4.26). Using the
above transformation for the root weight loss response, a simpler model was built
excluding four way interactions and the two models were compared. Significant
difference between the models was observed, so the full model was selected in place of
the simpler one. Since it was a full model with all main effects significant, the interaction
plots of all the factors need be discussed

154
Fig 4.24 Effect of pulling force on different medium and age

155
 Residuals vs Fitted Scale-Location
107 45

2.0
3

45

St anda rdi zed residuals

43
2

1.5
1
Residuals

1.0
-1

0.5
-2

43
-3

107

0.0
0 10 20 30 40 0 10 20 30 40

Fitted values Fitted values

Constant Leverage:
Normal Q - Q Residuals vs Factor Levels

45 45
4

4
Standardized residuals

Standardized residuals

2
2

0
0

-2
-2

-4

43
-4

43 107
107
Medium :
-3 -2 -1 0 1 2 3 M3 M2 M1

Theoretical Quantiles Factor Level Combinations

Fig 4.25 Diagnostics on model of plug weight loss due to fall (tried model)

156
 Residuals vs Fitted Scale-Location

3.0
107
130 108
0.5

130 108

2.0
0.0
Residuals

-1.5 -1.0 -0.5

1.0
107

0.0
0 1 2 3 4 5 6 0 1 2 3 4 5 6

Fitted values Fitted values

Constant Leverage:
Normal Q - Q Residuals vs Factor Levels
5

108 130 130 108

Standardized residuals

Standardized residuals

0
0

-5
-5

107
-10

107
Medium :
-3 -2 -1 0 1 2 3 M3 M2 M1

Theoretical Quantiles Factor Level Combinations

Fig 4.26 Diagnostics on model of plug weight loss due to fall - model after square
root transformation

157
4.5.1 Interactive effect of falling height and cushion thickness

Figure 4.27.a shows that the effects of falling height and cushion thickness on plug
root weight loss. It is noted that the root weight loss was found to be lesser
(5 per cent) with cushion of 40 and 20 mm at 250, 500 and 750 mm falling height. If the
20 mm cushion thickness was used, the root weight loss was 5 per cent and 5 to 10 per
cent at 250, 500 and 750 mm falling height respectively. The root weight loss increased
from 10, 15 and 20 to 25 at 250, 500 and 750 mm falling height for treatment without
cushion. Quite obviously, as the cushion thickness increased, the damage to root ball
decreased. Similarly as the height of fall increased, the loss was higher due to the higher
impact. The trend shows that the damage would be linearly increasing further, if the
height of fall is more than 750 mm. The rate at which the loss occurs was not uniform for
the provision of cushion. The difference in loss between 20 and 40 mm cushions was
lesser than that between 0 and 20 mm. The provision of 20 mm cushion would then be
considered optimal. Since more thickness would not be practically feasible in the
machine.

4.5.2 Interactive effect of cushion thickness and plug medium

It is observed from figure 4.27.b that the coir pith alone when used as growth
medium gave the maximum (25 per cent) plug root weight loss as found to be without
cushion. Subsequently the plug root weight was reduced (8 per cent) in 40 and 20 mm
cushion thickness. In the coir pith + clay soil (2 per cent by volume) and coir pith + clay
soil (5 per cent by volume) growth mediums, the plug root weight loss was minimized
with cushion of 20 and 40 mm (less than 5 per cent) and without cushion (10 per cent).
The root weight loss was reduced because the clay soil mixture firmly held the media and
roots together without affecting the germination of the seeds as well as the
reestablishment of the seedling in the field.

Hence either the coir pith + clay soil (2 per cent by volume) or coir pith + clay soil
(5 per cent by volume) could be used as media for reduced weight loss due to fall. The
clay soil mixture increased the weight of root ball, thus reducing the straightening height
of plant which was proven in the subsequent weight loss experiments.

158
 The loss was acute in the case of coir pith alone, but the addition of clay soil at
2 or 5 per cent did not make much difference. This has been indicated in the Tukey test
(Fig.4.30) mean comparison, where the clay soil mixed media show on par loss in root
ball weight while dropped.
Per cent loss of plug medium
25

Cushion thickness
20

0 mm
20 mm
15

40 mm
10
5

250 500 750

Height of fall, mm

a. Effect of cushion and height of falling

Per cent loss of plug medium

25

Cushion thickness
20

0 mm
15

20 mm
40 mm
10
5

Coi pith Coir pith + 2% clay Coir pith + 5 % clay

Plug medium

b. Effect of cushion and plug medium

Fig 4.27 Influence of cushion thickness, media and height of fall on weight loss due
to fall

159
4.5.3 Interactive effect of hardening and cushion thickness

Figure 4.28.a illustrates the effect of hardening and cushion thickness on root
weight loss. It was found that the minimum (5 per cent) root weight loss was observed
without hardening and with cushion of 40 mm thickness. After one day hardening the
root medium became more brittle, hence the loss was increased. The same tendency was
observed in 20 mm cushioning also. The root weight loss (5 to 10 percent) was drastically
reduced at 1.5 and 2 day of hardening, because the moisture content of the medium dried
up and the medium becomes solid. However it was observed that if the seedling was
dried over 2 day hardening, it may affect the survival of the seedling in the field after
transplanting although the loss was minimal in all cases. The maximum root weight loss
was (20 to 25 per cent) as found without cushion because of the impact. The Tukey mean
comparison chart (Fig.4.30) clearly indicated that the root ball weight loss due to fall,
was on par in 0 and 1.5 day of hardening and hence it was proven that the hardening of
nursery is really not required and can be directly transplanted to reduce the plug weight
loss due to fall.

4.5.4 Interactive effect of plug medium and height of falling

The coir pith + clay soil 5 per cent (by volume) and coir pith + clay soil 2 per cent
(by volume) recorded minimum (less than 2 per cent) weight loss. The presence of clay
soil had improved the solidity of the root ball (Fig.4.28.b). The interaction between the
height of fall and media was interesting that unlike the effect of cushion thickness, the
increase in height of fall increased the plug weight loss uniformly. Tukey mean
comparison (Fig.4.30) shows that the weight loss of different heights of falls are
significantly different from each other.

4.5.5 Interactive effect of hardening and height of falling

Figure 4.29.a shows a definite inflexions of curves at 1.5 day of hardening for 500
and 250 mm heights of falls, which was absent for 750 mm fall. This influencing made
the loss per cent equal for 0 and 1.5 day of hardening for 500 and 250 mm falls. But
practically at least 750 mm of fall height would be required for the plants to straighten
and hence the above influence may not be relevant to the machine.
160
4.5.6 Interactive effect of plug medium and day of hardening

Figure 4.29.b shows once again that the effect of media with soil mixture (2 per
cent or 5 per cent by volume) did not make any difference in shattering of plug.
Per cent loss of plug medium

25

Cushion thickness
20

0 mm
20 mm
15

40 mm
10
5

0.0 1.0 1.5 2.0

Days hardening

a. Effect of cushion and hardening

Per cent loss of plug medium

25

Height of fall
20

750 mm
15

500 mm
250 mm
10
5

Coi pith Coir pith + 2% clay Coir pith + 5 % clay

Plug medium

b. Effect of height of falling and plug medium

hardening

Fig 4.28 Influence of media, day of hardness and cushioning on plug weight loss due
to fall

161
 But it was observed that 2 day hardening had made the 5 per cent soil mixed media
real hard which had reduced the loss to negligent levels. But the survival of the 2 day
hardened nursery needs to be verified further in the field after transplanting. The 1.5 day
hardened nursery gave a marginal decrease in loss as compared to without hardening.

Model listing III shows the sequence of statistical script modelling weight loss
data.
Per cent loss of plug medium
25

Height of fall
20

750 mm
15

500 mm
250 mm
10
5

0.0 1.0 1.5 2.0

Days of hardening
a. Effect of hardening and height of falling
Per cent loss of plug medium

25

Days of hardening
20

0.0
15

1.0
1.5
10

2.0
5

Coir pith Coir pith + 2% clay Coir pith + 5 % clay

Plug medium

b. Effect of hardening and plug medium

Fig 4.29 Effect of media, day of hardening and height of fall as weight loss due to
falls

162
 M2-M3

D1-D4
M1-M3

D1-D3
M1-M2

D2-D1
0.5 1.0 1.5 2.0 2.5 0.0 0.2 0.4 0.6 0.8 1.0

Differences in mean levels Differences in mean levels

as influenced by growth medium as influenced by days of hardening
H2-H3

C2-C3
H1-H3

C1-C3
H1-H2

C1-C2

0.6 0.8 1.0 1.2 1.4 1.6 0.5 1.0 1.5 2.0
Differences in mean levels Differences in mean levels
as influenced by fall height as influenced by cushion thickness

Fig 4.30 Tukey mean comparison on the plug weight loss due to fall

Growth medium Hardening day

M1 - Coir pith D1 - Without hardening
M2 - Coir pith + clay soil (2 per cent) D2 - 1.0 day hardening
M3 - Coir pith + clay soil (5 per cent) D3 - 1.5 day hardening
D4 - 2.0 day hardening
Falling height Cushion thickness
H1 - 500 mm C1 - Without cushion
H2 - 750 mm C2 - 20 mm
H3 - 1000 mm C3 - 40 mm

163
 Model listing - III

Modeling root ball weight loss due to fall

> #Reading the data in

dat1=read.table("e:/download/SivakumarProject/weightloss.csv", sep=',', header =

TRUE)

> #Plug weight loss analysis

> #Trying the full anova model

> a1=aov(loss~medium*harden*height*cushion, data = dat1)

> summary(a1)

Df Sum Sq Mean Sq F value Pr(>F)

medium 2 16028 8014 12612.35 <2e-16 ***

harden 3 1990 663 1044.05 <2e-16 ***

height 2 3852 1926 3031.49 <2e-16 ***

cushion 2 11484 5742 9036.87 <2e-16 ***

medium:harden 6 1093 182 286.62 <2e-16 ***

medium:height 4 1016 254 399.58 <2e-16 ***

harden:height 6 267 45 70.07 <2e-16 ***

medium:cushion 4 1614 403 634.99 <2e-16 ***

harden:cushion 6 1184 197 310.51 <2e-16 ***

height:cushion 4 772 193 303.76 <2e-16 ***

medium:harden:height 12 217 18 28.48 <2e-16 ***

medium:harden:cushion 12 2689 224 352.61 <2e-16 ***

medium:height:cushion 8 314 39 61.70 <2e-16 ***

harden:height:cushion 12 298 25 39.14 <2e-16 ***

164
medium:harden:height:cushion 24 697 29 45.70 <2e-16 ***

Residuals 216 137 1

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> #Open another graphical window and split screen into 2 by 2 areas

> windows()

> layout(matrix(c(1,2,3,4),2,2))

> #Diagnostics plot

> plot(a1)

> #Normalcy assumption is not obeyed

> #Going for Box cox transformation test

> library(MASS)

> windows()

> #Drawing box cox diagnostics

> boxcox(a1)

> windows()

> #Drawing box cox with more resolution

> boxcox(a1,lambda = seq(0.5, 0.8, 0.05))

> #Applying the lamda transformation to response variable

> #modeling in full

> a1=aov(loss*(0.65)~medium*harden*height*cushion, data = dat1)

> summary(a1)

Df Sum Sq Mean Sq F value Pr(>F)

medium 2 6772 3386 1 2612.35 <2e-16 ***

165
harden 3 841 280 1044.05 <2e-16 ***

height 2 1628 814 3031.49 <2e-16 ***

cushion 2 4852 2426 9036.87 <2e-16 ***

medium:harden 6 462 77 286.62 <2e-16 ***

medium:height 4 429 107 399.58 <2e-16 ***

harden:height 6 113 19 70.07 <2e-16 ***

medium:cushion 4 682 170 634.99 <2e-16 ***

harden:cushion 6 500 83 310.51 <2e-16 ***

height:cushion 4 326 82 303.76 <2e-16 ***

medium:harden:height 12 92 8 28.48 <2e-16 ***

medium:harden:cushion 12 1136 95 352.61 <2e-16 ***

medium:height:cushion 8 133 17 61.70 <2e-16 ***

harden:height:cushion 12 126 11 39.14 <2e-16 ***

medium:harden:height:cushion 24 294 12 45.70 <2e-16 ***

Residuals 216 58 0

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> windows()

> layout(matrix(c(1,2,3,4),2,2))

> #Diagnostics plot again

> plot(a1)

> # diagnostics still bad

> #Going for square root tranformation

> a1=aov(sqrt(loss)~medium*harden*height*cushion, data = dat1)

> summary(a1)
166
 Df Sum Sq Mean Sq F value Pr(>F)

medium 2 446.4 223.21 6649.777 < 2e-16 ***

harden 3 34.5 11.50 342.689 < 2e-16 ***

height 2 141.8 70.89 2111.936 < 2e-16 ***

cushion 2 315.2 157.58 4694.649 < 2e-16 ***

medium:harden 6 24.0 4.00 119.228 < 2e-16 ***

medium:height 4 3.2 0.80 23.876 2.29e-16 ***

harden:height 6 5.0 0.84 24.990 < 2e-16 ***

medium:cushion 4 28.6 7.14 212.733 < 2e-16 ***

harden:cushion 6 23.3 3.88 115.619 < 2e-16 ***

height:cushion 4 0.2 0.05 1.592 0.178

medium:harden:height 12 4.0 0.33 9.891 2.47e-15 ***

medium:harden:cushion 12 62.3 5.19 154.682 < 2e-16 ***

medium:height:cushion 8 6.0 0.75 22.300 < 2e-16 ***

harden:height:cushion 12 6.9 0.58 17.173 < 2e-16 ***

medium:harden:height:cushion 24 12.6 0.52 15.585 < 2e-16 ***

Residuals 216 7.3 0.03

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> windows()

> layout(matrix(c(1,2,3,4),2,2))

> #Diagnostics plot again

> plot(a1)

> #Diagnostics better now

> #Building a simpler model excluding four way interaction

167
> a2=aov(sqrt(loss)~(medium+harden+height+cushion)^3, data = dat1

> #Comparing the two maodels

> anova(a1,a2)

Analysis of Variance Table

Model 1: sqrt(loss) ~ medium * harden * height * cushion

Model 2: sqrt(loss) ~ (medium + harden + height + cushion)^3

Res.Df RSS Df Sum of Sq F Pr(>F)

1 216 7.2504

2 240 19.8054 -24 -12.555 15.585 < 2.2e-16 ***

Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

> #Significant difference between the models, so goiing with the full model

> #Open yet another gaphical window and split screen into 2 by 2 areas

> layout(matrix(c(1,2,3,4),2,2))

> #Plot the tukey test plots of all factors

> #Whichever range of mean diffeneces between pairs cross zero are on par

> plot(TukeyHSD(a1, "medium", or

> plot(TukeyHSD(a1, "height", ordered = TRUE))

> plot(TukeyHSD(a1, "harden", ordered = TRUE))

> plot(TukeyHSD(a1, "cushion", ordered = TRUE))

> #Drawing the interaction plots of all factors to visualize the means

> windows()

> layout(matrix(c(1,2),2,1))

> with(dat1,interaction.plot(height,cushion,loss, lwd=2))

> with(dat1,interaction.plot(medium,cushion,loss, lwd=2))

168
> windows()

> layout(matrix(c(1,2),2,1))

> with(dat1,interaction.plot(harden,cushion,loss, lwd=2))

> with(dat1,interaction.plot(medium,height,loss, lwd=2))

> windows()

> layout(matrix(c(1,2),2,1))

> with(dat1,interaction.plot(harden,height,loss, lwd=2))

> with(dat1,interaction.plot(medium,harden,loss, lwd=2))

169
4.6 Logit modeling of experiments on straightening height of seedling in free fall

The experiments on straightening height of the seedling were explained in section

3.7.3. The influencing factors considered were the growth media at levels of coir pith,
coir pith + clay soil 2 per cent (by volume) and coir pith + clay soil 5 per cent (by
volume) and the height of falls as 250, 500 and 750 mm. The probability of the success or
failure of straightening of seedling at different height were observed. The logit modelling
procedure described in section 4.2 was followed to optimize the straightening height. The
generalized linear model was developed as a full model with interaction of medium and
height. The built model listing is shown in the model listing - I. The model was found to
be adequate as seen from the AIC. To simplify the model, the model was updated without
any interactions and comparing these two models, the model without any interaction was
found to be sufficient. Hence the simpler model was found to be optimal model as to
contain height alone. The predicted values of the built model are shown as lines and
observed probability as points (Fig.4.31).

Figure 4.31 illustrates that the predicted values are very close to the observed
values at all the heights. It was inferred that the probability of straightening was 70 to 80
per cent at 500 and 750 mm height for the different medium. If 750 mm height was
chosen for straightening of the seedling, the root ball weight loss due to impact on fall
may be excessive (Brewer, 1997). Hence 500 mm height was considered optimal.
Although the experiment showed only a maximum of 80 per cent straightening success,
this observation was for free fall. But when the fall happens in the enclosed hopper it
increased to more than 90 per cent.

4.7 Optimization of machine parameters influencing vegetable seedling

transplantation performance

As discussed in section 3.9, a field test was made to optimize the relevant
parameters involved in the vegetable seedling transplantation process. The parameters
were the forward speed, planting depth (furrow depth), field moisture content and
compaction wheel angle. The response was the uprightness of the transplanted seedling

170
and was considered binary with success/ failure as levels. A logit model was built on the
planting response as influenced by the said factors.

1.0
Probability of successful uprights
0.8
0.6
0.4
0.2
0.0

250 500 750

Height of fall, mm

Fig 4.31 Effect of height of fall on success of straightening

Therefore after modeling the experimental data and testing it for adequacy and
simplicity, predictions were carried out based on the developed logit model. The
predictions gave the probability of success of plant uprightness after transplanting. Since
the RHS of the logit model is nothing but a linear model, the fitting of model was started
from full model inclusive of interaction between the independent variables and finally
simplified to an adequate model. Using the model, one could predict the plant uprightness
of the transplanted seedling as influenced by the forward speed, planting depth, moisture
content of the field and compaction wheel angle.

171
 The seedling uprightness and proper coverage and compaction of the root ball was
considered as the transplanting performance of the machine. The probability of success in
transplanting the seedling in upright position was hence taken as an index of
transplanting performance.

4.7.1 Transplanting performance at a forward speed of 0.5 km h-1

Figure 4.32 indicates that the moisture content and compaction wheel angle
influence the probability of successful transplantation at a depth of 50 mm. The
compaction wheel angle of 15 with moisture content of 10 per cent has about
60 per cent probability of success and the 20 per cent moisture content with 0 
compaction wheel angle has shown a poor response with the probability of successful
uprightness moving down to 20 per cent. An optimum wheel angle of 15 at the least
moisture of 10 per cent had provided good performance. When the planting depth was
increased to 75 mm, the performance was quite similar in terms of the relationships.
However, the optimal compaction wheel angle has performed much better at 10 per cent
soil moisture with about 80 per cent successful transplantation. The decrease in
performance with moisture was in the same rate as that in 50 mm depth of operation.
However at the 75 mm depth of operation and 30 compaction wheel angle performed the
poorest of all. When the depth of planting was taken up to 100 mm, the transplanting
performance deteriorated much. The maximum success probability was only about 40 per
cent. (Satpathy and Garg, 2008; Narang et al., 2011; Nambu and Tanimura 1992;
Rumsey, 1989; and Kumar and Raheman, 2011).

172
 1.0
0 deg wheel angle
15 deg wheel angle

Probability of successful planting

30 deg wheel angle

0.8
0.6
0.4
0.2
0.0

10 15 20
Soil moisture, per cent
a. Depth 50 mm
1.0

0 deg wheel angle

15 deg wheel angle
Probability of successful planting
0.8

30 deg wheel angle

0.6
0.4
0.2
0.0

10 15 20
Soil moisture, per cent

b. Depth 75 mm
173
 1.0
Probability of successful planting
0.8
0.6
0.4
0.2
0.0

10 15 20
Soil moisture, per cent
c. Depth 100 mm

Fig 4.32 Probability of successful uprightness due to the effect of compaction wheel
angle and depth of planting at forward speed of 0.5 km h -1
4.7.2 Transplanting performance at operating speed of 1 km h-1 forward speed

The influence of soil moisture and compaction wheel angles are exactly similar to
these at a slower speed of 0.5 km h -1 (Fig 4.32a, 4.32b and 4.32c.). However the success
of transplantation has gone down drastically due to higher speed. As against a maximum
probability of success of 80 per cent at slower speed, here it was only 60 per cent
occurring at the 15 compaction wheel angle and at 10 per cent soil moisture (Fig 4.33).
As the speed of travel increased, the seedling that was dropped had lesser time to position
itself in the furrow before the soil heaped by the compaction wheels, move into bury the
root ball. The most important factor was the horizontal velocity imparted by the machine
to the dropped seedling, which made it to lean forward and caused it to be buried by the
flowing soil. The increase in depths from 50 mm to 75 mm had improved performance
but depth above 75 mm caused a drop in performance.

174
 1.0
0 deg wheel angle
15 deg wheel angle

0.8
30 deg wheel angle

Probability of successful planting

0.2 0.4 0.6
0.0

10 15 20
Soil moisture, per cent
a. Depth 50 mm
1.0

0 deg wheel angle

15 deg wheel angle
30 deg wheel angle
0.8
Probability of successful planting
0.2 0.4 0.6
0.0

10 15 20
Soil moisture, per cent

b. Depth 75 mm
175
 1.0
Probability of successful planting
0.8
0.6
0.4
0.2
0.0

5 10 15
Soil moisture, per cent

c. Depth 100 mm

Fig 4.33 Probability of successful uprightness due to the effect of compaction wheel
angle and depth of planting at forward speed of 1 km h -1
4.7.3 Transplanting performance at operating speed of 1.5 km h-1 forward speed

The general conclusion that the forward speed of travel deteriorates performance
was proved further from the data at forward speed of 1.5 km h-1. The success of
transplantation reduced to 50 per cent maximum at 10 per cent soil moisture and 15
wheel angle. Otherwise all the other relationships remained similar to the results for the
slower speeds. At 100 mm depth of placement, the curves for the three levels of
compaction wheel angles overlapped and was showing a very poor performance with
only 10 to 20 per cent success rate (Fig 4.34).

The results on transplanting performance of the machine showed that, the optimum
moisture content of soil is 10 per cent (w.b), the wheel angle 15 and the speed 0.5 km h
-1
.

176
 1.0
0 deg wheel angle
15 deg wheel angle
30 deg wheel angle

0.8
Probability of successful planting
0.2 0.4 0.6
0.0

10 15 20
Soil moisture, per cent
a. Depth 50 mm
1.0

0 deg wheel angle

15 deg wheel angle
30 deg wheel angle
0.8
Probability of successful planting
0.2 0.4 0.6
0.0

10 15 20
Soil moisture, per cent
b. Depth 75 mm
177
 1.0
0.8
Probability of successful planting
0.6
0.4
0.2
0.0

10 15 20
Soil moisture, per cent
c. Depth 100 mm

Fig 4.34 Probability of successful uprightness due to the effect of compaction wheel
angle and depth of planting with forward speed of 1.5 km h -1
4.8 Endurance test on the developed machine

An endurance test was conducted on the developed machine which revealed that for
every 120 to 140 punches, the plunger was not able to align perfectly to the protray hole
and after every 300 to 400 punches the motor was damaged and the motor needs to be
replaced. The fabrication of the machine should be done more preciously in future with
appropriate replacements of good quality geared motors and mechanical links. A fine
pitched screw and nut movement may be thought of instead of the walking beam
mechanism, which proved to be not very reliable in the field machines.

4.9 Field performance of the developed automatic transplanter

The prototype automatic vegetable transplanter was evaluated in the field for
transplanting tomato seedling with the optimal value of compaction wheel angle at 15 
as set on the automatic vegetable transplanter machine. During the final field trials on the

178
automatic vegetable transplanter, a speed of 0.5 km h-1 and a soil moisture content of 10
per cent were ensured to bring about the best performance. It was found that the
transplanter placed the seedling accurately in the furrow at the expected spacing
(450 mm) with a variation of ± 25 mm with the success ratio of 80 per cent. The plant
missing, plant mortality after 20 days, the depth of planting and planting angle were
observed. The plant missing was well within the acceptable limit of 3 - 4 per cent and
plants in laying down position and plant mortality were lower (10 per cent) with the
percentage of upright plants highest (82 per cent). The soil covering efficiency of the
automatic vegetable transplanter was found to be about 86 per cent. The field capacity of
-1
the machine was found to be 0.02 ha h . The plants survived well after being
transplanted in the field and provided a better field stand.

179
Summary and Conclusions
 CHAPTER V

SUMMARY AND CONCLUSIONS

India has emerged as a leading horticultural country of the world with a total annual
production of 240 million tonnes of horticultural crops during 2010 - 11. The
transplanting operation is one of the most labour intensive in vegetable production. It is
largely done manually in India and incurs large investments in labour, time and cost.
Each field worker is only able to set 400 plants h-1 (Suggs et al., 1989), so a considerable
labour force is demanded. While transplanting is done manually, roots are severely
damaged in the process of transplanting, so the plants take longer to establish their roots
after transplanting.

Plug or cell transplants are seedling or small vegetatively propagated plants which
are raised in individual small cells, called plugs. Plug trays or protrays are available in
plastic, which have an array of cells. The plugs are filled with a cohesive medium, plants
grown, and are eventually transplanted into other growing systems. Vegetable crop
growers simply buy the transplants of their choice from the plug growers and devote their
efforts to growing their crops rather than nursery. Plug plants are uniform and high in
quality. Lot of entrepreneurs have put up net houses, exclusively for supply of protray
grown seedlings.

The currently developed transplanters in India are semi automatic transplanters.

They are not suitable for continuous operation over a long period of time. Here the
feeding of seedling is done manually and metered mechanically. The field capacity of
these units is however restricted by the feeding capacity of labour. The transplanter needs
to be operated at a very slow speed so that the operator will have time to pick the seedling
and drop it for transplanting.

In developing an automatic vegetable transplanter, the foremost factor is the

seedling removal which needs to be automated, where seedlings need be removed from
the protray one by one without damage to the seedling stem or root balls. The mechanism

180
of ejection is intricate and two methods were contemplated, one where the plug is
released by weight of the root ball and the other ejected by force.

The protray seedling instead of being drawn from the top was planned to drop
through the bottom by gravity. Open bottom protrays are not available commercially. So
protray cell bottoms were manually cut and used on trial basis, instead of using the whole
protrays. Protray was segmented into seven linear strips and used, so that each strip with
seedling can be fed to the machine rather than the whole tray. The protray strips were cut
manually to uniform size and the bottom was removed manually. The protray segment
was placed upside down, media filled and seeds sown in its bottom - up position. When
the seedlings are fully grown the taper available in the cell allows the seedling to slide
down easily by gravity.

An experimental feeding chute was developed to test the process by feeding the
protray strips towards testing whether gravity ejection is really feasible. A 12 V, DC,
50 rpm centre shaft gear motor with a finger wheel was used for providing the movement
to the seedling segment. The mechanism was not able to smoothly move the protray
segment, since the fingers were pushing the cells tangentially.

So a walking beam mechanism using double crank with belt drive was designed to
provide a perfect axial motion to the tray cells. The twenty days old seedlings were tested
for free fall on the test rig developed. The results showed that most of the plants did not
successfully fall down, since the media stuck to the protray surface and gravity was not
sufficient to release them. To overcome this problem, it was planned to try different
media mixtures and also to use a paper cone within the tray cell, so that the plugs can
release from the cells easily by gravity.

It was observed that the seedlings grown in soil + vermicompost (4:1) and in paper
cone fell down successfully due to the higher weight of the seedling. However, this
method was found not implementable in the contemplated machine, due to problems in
properly feeding the segments, the non availability of precisely cut protray segments and
older seedlings not falling through the plug cell.

181
 A four bar mechanism with an appropriate coupler path was designed to extract the
seedling from the tray cell from the top of the tray and release it to fall free to the ground.
The tip of the claw, as an extension of the coupler bar, was supposed to extract the plug
seedling entering into the top of the protray cell and tease it out, by moving in a
predefined path. A tray holder was developed wherein the protray with seedling could be
placed vertically. The pick - up claw was expected not to break up the root ball during
removal. The protray seedling removal was successful but the root ball being fragile often
broke up. This method of ejection was hence discarded.

A hand cranked slider crank mechanism was then devised to pop out the seedling as
pushed from the bottom. The bottom of a conventional 14 x 7 protray was modified with
a 10 mm hole at the cell bottoms. The seedlings were grown in these trays by using
conventional growing medium for twenty five days, for experiments with this method.
A 6 mm round rod was used as a plunger for pushing out the seedling from the bottom of
the protray. The rod entered to a depth of 80 per cent of the protray cell depth so that the
seedlings were positively ejected out. Preliminary trials were found to be satisfactory, so
the ejection of the seedling by popping from back was decided to be motorised.

A 12 V, 100 rpm, DC geared motor was used to operate the slider crank
mechanism. The popping out of the seedlings was to be done from bottom row first
upward so that the ejected plant could fall down freely without entanglement. But when
the plunger motor was operated to move the plunger forward and retract, switching off
the motor after one such stroke may not instantly stop the motor and so the plunger. Due
to inertia, the plunger would move a bit further and stop at some random position inside
the protray cell, thus obstructing the linear movement of the protray itself for extracting
the other cells using the same plunger. Hence it was conditional that the motor should be
stopped instantly after the plunger made one stroke ejecting the plug and retracting the
plunger back. To achieve this, the principle of regenerative braking was used and a circuit
with appropriate switching was successfully developed to precisely stop the retracted
plunger.

182
 A statistically robust experiment was planned to assess the ejection performance of
this method. The four selected independent variables were age of seedling, growth
media, plunger diameter and the speed of plunger. The success of ejection or failure was
recorded in each replication and logit analysis was attempted on the generated data.

The predicted values by the statistical model perfectly fitted the observed
probability values and for the 6 mm diameter plunger, the probability of successful
ejection was more than 90 per cent for all the media. While using 5 and 4 mm plunger
diameter for ejection, the probability of success was about 80 to 85 per cent in all the
media. But the 3 mm plunger diameter instead of ejecting, penetrated the medium and
therefore the probability of ejection was less than 80 per cent.

A 75 per cent shade net house was exclusively built for raising plug transplants
with different growth medium. The main difficulty faced in growing the protray seedling
was due to the provision of large hole at the bottom of the cells. There was no problem of
retaining the media in, because it was compacted into the cells before sowing. When the
open bottom trays are placed on plastic covered platforms, the roots propagated out of the
tray and caused problems in ejection. However when the media loaded trays were placed
on racks with their bottoms open to air, the roots did not propagate out of the tray. A rack
system hence was developed for growing the plug transplants in the constructed shaded
house. Further experiments were done using these rack grown seedlings.

The seedling ejection from the protray alone is not adequate, but the plant itself
should grow healthily for transplanting in the selected medium that provided good
ejection. Inadvertently designing the media for machine suitability would not be
acceptable. So to assess the plant growth as influenced by age and growth medium, they
were experimented in terms of plant height, stem diameter and number of leaves.

The analysis of the result was expected to show whether the growth is affected by
the change in media and age. In addition to the agronomic health of the plants, the root
plug weight and stem weight were also measured and analysed. The growth medium was
washed off and wet weight of root was also recorded. Unconventionally, soil was also
tried as mixed with coir pith to ascertain the agronomic influence. 40 days old seedlings

183
as grown in coir pith + vermicompost 4:1 reached the maximum height of 230 mm. The
20 days old seedling grown in coir pith + soil (1:1) grew only to a height of 110 mm. The
stem thickness was quite high at 3.6 mm in 40 days old seedlings grown in coir pith +
vermicompost (4:1). The minimum stem thickness (2.4 mm) was found in 20 days old
seedling in coir pith alone. The number of leaves was found to be high in 40 days old
seedlings grown in coir pith + vermicompost (4:1) and the minimum number of leaves in
20 days old seedling grown in coir pith alone. Both the age and type of medium were
significantly influencing the growth of seedling.

The age of seedling was found to be non significant on the plug weight. The plug
weight was found to be more in coir pith + sand + clay soil (17.86 g) and the lowest
(10.68 g) for coir pith in the 20 days old seedling. The root weight was highest (0.5 g) as
grown in coir pith + vermicompost (4:1) and lowest was in coir pith + soil (1:1) (1g) on
20 days old seedling. The experiment on the growth parameter provided information on
whether the medium and age can be altered to suit machine condition. Coir pith with
vermicompost was proven superior to all other media combination in this experiment.

The pulling force was measured to assess the ease of ejecting out the seedling form
the protray as influenced by the growth medium and age of seedling. The media with soil
compositions had more pulling force (1.6N) than coir pith + vermicompost (1.03N). This
is due to the soil colloids which stick on to the protray.

After optimizing the plug ejection performance, it was contemplated that the
ejected seedling, when falling down to the ground directly or on a trap door during
transplanting, may get damaged in the root ball. To determine the weight loss of root ball
of the optimized medium (Coir pith) while falling down, a drop test was conducted at
different falling heights of 1000, 750, 500 and 250 mm. The seedlings used for testing
was either without hardening, one day, one and half day or two day hardening.

The growth medium was altered with 2 or 5 per cent sundried black soil with coir
pith is an effort to reduce the root ball disintegration due to impact. A layer of cushion of
variable thickness was also added to the experiment as a variable. A planned experiment
was done on the plug weight loss on fall as influenced by media, cushion thickness and

184
falling height. The seedling was dropped down manually on the surface without or with
a sponge cushion of 20 and 40 mm thickness.

The root weight loss was found to be lesser (5 per cent) with a cushion of 40 and
20 mm at 250, 500 and 750 mm falling height.

In the contemplated transplanter, it was planned that the seedling has to make a free
fall to the ground after ejection. Hence it is imperative that the horizontally ejected
seedling should straighten when it reached the ground level.

The straightening height of the seedling was found, which was used for designing
the hopper in the vegetable transplanter. A hard board cone was fixed on a rectangular
frame, through which the seedling will fall down and arrangement was provided to fix the
frame at different heights of 250, 500 and 750 mm, from the ejection level.

A prototype vegetable transplanter was developed based on the optimized

variables obtained from the experimental data. The prototype consisted of a main frame,
hopper for guiding the dropped seedling to ground, furrow opener and compaction wheel.
The furrow opener cuts a narrow furrow of 60 mm into which the seedlings dropped by
the hopper fall into in an upright position between the set of rolled/ yawed compaction
wheels. The soil is so heaped and compacted around the root ball making the plant to be
transplanted. The seedling that falls though the hopper is checked by a trap door/valve at
the bottom end of the hopper. The trap valve was cushioned on the top to reduce the root
ball shattering and opened by a trip mechanism activated by the ground wheel. The
trailed ground wheel trips the valve every 450 mm of travel, to plant the seedling at the
desired plant spacing on the row.

A field optimization experiment was done to determine the influence of the

transplanting variables on planting performance. The four selected variables were speed
of travel, planting depth, compaction wheel angle (yaw) and soil moisture content as
influencing the field performance of the vegetable transplanter. The seedling uprightness
and proper coverage and compaction of the root ball was considered as the transplanting
performance of the machine. The above prototype did not have any provisions for
automatic feeding of seedlings. So the seedlings were manually dropped from a height of
185
500 mm into the hopper at regular intervals of 2 seconds so as to test the performance.
Since it was manual method, the transplanted spacing was not taken as response. The
results on transplanting performance of the machine yielded that, the optimum
moisture content of soil is 10 per cent (w.b), the wheel angle 15 and the speed of
travel 0.5 km h -1.

Having optimized the field performance parameters of the prototype transplanter,

an automatic transplanter mechanism was devised next to automatically pop out the
seedlings from the seedling tray and plant them in the furrow.

The automatic transplanting mechanism had the following constituent parts.

1. A horizontal rail and slider assembly carrying the vertically loaded seedling
protray, which slides along the travel direction of the machine

2. A vertical column of seven ejector mechanisms that are positioned so as to eject

plugs from one single vertical column of the protray at a time

3. A motorized walking beam mechanism to move the tray slider in precise steps
corresponding to the inter distance between cells on the protray.

4. Optical detector for precise positioning of the pro-tray cells relative to the ejector
plungers.

5. Main microcontroller (PIC16F128) board controlling the logic of tray movement

and plug ejections in proper sequence.

6. An exclusive relay board for driving the seven ejector motors along with
hardware for regenerative stopping of these motors.

The horizontal protray slider assembly was to move the protray axially. It had a
protray holder which holds the protray in vertical position.

Seven ejector motors were arranged one over the other in a vertical column. Each
ejector assembly had a plunger driven by a geared motor and a limit switch for exactly
stopping the plunger retracted after completion of one popping cycle. To avoid tangling
of seedlings during ejection, the seedlings were ejected from bottom to top of the protray.

186
 The motorized walking beam mechanism provides a linear movement of the slider,
which moved the slider with the protray laterally to the ejectors by 36 mm.

The optical detector was added to this arrangement additionally to precisely stop
the moving tray at 36 mm spacing as controlled by the microcontroller.

The main microcontroller board provided the logic for the automatic sequence of
operation of the machine. The microcontroller accepted a switched signal from the
ground wheel for detecting the plant spacing on the row as well the signal from the opto
sensor for precisely positioning of the protray cell centre with respect to the ejector
plungers.

When the activation reached the microcontroller from the ground wheel spacing
detector, it started to sequentially operate seven ejectors from bottom to top thus ejecting
one column of seedlings from the protray. Once that operation was over, the
microcontroller issued drive command to the walking beam mechanism to move the tray
to the next column. The ejection of that column was invoked by the controller and so on,
till all the 14 columns of the seedling tray was emptied. Once the tray was empty, that is
after a travel of 98 x 0.45m  44m of the planter, it was stopped and a fresh seedling tray
was loaded. Now the microcontroller starts the ejection sequence of the last column of
the newly fed tray. After the sequence, the controller reversed the polarity to the walking
beam drive and moved the protray in the opposite direction so that the previous to last
column can be ejected and so on. This ejection/ movement sequence was completely
controlled by the microcontroller. The program code for the explained events was
embedded in the PIC memory for operation.

After the incorporation of the automatic transplanting mechanism on the prototype

transplanter, a field test was done to assess the performance of the machine. The field was
prepared by rototilling it twice at a soil moisture of 10 per cent wet basis. The prototype
machine was operated at a speed of 0.5 km h-1, set to plant at 75 mm depth with compaction
wheel angle set at 15 . The plants angled at less than 30  with the horizontal plane after
transplanting were considered as buried. 20 days old tomato seedlings with plant height of
120 to 140 mm were used for field evaluation.
187
 It was found that the transplanter placed the seedling accurately in the furrow at the
expected spacing (450 mm) with a variation of ± 25 mm and a success ratio of 80 per cent.
The plant missing , plant mortality after 20 days, depth of planting, planting angle,
number of plants lying down were noted. The plant missing was well within the
acceptable limit of 3 - 4 per cent. Plant mortality was lower (10 per cent) with the
percentage of upright plants being highest at 82 per cent. The soil covering efficiency of
the automatic vegetable transplanter was found to be about 86 per cent. The field capacity
-1
and field efficiency of the machine were found to be 0.015 ha h and 50 per cent. The
cost of operation of the automatic vegetable tranplanter was Rs.2072 ha-1, which was less
than conventional method. The transplanter resulted in 97 per cent saving in labour and
90 per cent in labour cost. The cost of seedling was Rs.23,187 ha-1.

An endurance test was conducted on the developed machine which revealed that
after every 120 to 140 punches the plunger was not able to align perfectly to the protray
hole and after every 300 to 400 punches the motor got damaged and the motor needs to
be replaced to make the machine work.

As a whole, this work enabled development of a working model of an automatic

vegetable transplanting machine which can eject the seedling from a protray and plant them.

However the prototype needs to be built more ruggedly to be evolved into a trouble
free machine. Work can be directed in future in the following lines.

* The walking beam mechanism can be replaced with a screw and half nut
mechanism for better operation of the machine.

* Servomotors can be used instead of geared DC motors to eliminate the need for
regenerative stopping circuits and the concerned hardware, subject to the
availability of small servos.

* The protray used for the automatic transplanter should be more rugged, solid and
with larger drain hole to avoid problem of popping accuracy. The protrays available
in the market are very flimsy and not suitable for machine handling. Exclusive
protrays could be fabricated to work with the machine.

* The uniform height of plug plants is also mandatory for transplant automation to
succeed.

188
References
 REFERENCES

Anonymous.1983. The techiniculture transplant system. Sales literature. Castle and

Cooke Techiniculture, Inc, 56 Peckham Rd, Watsonville, CA95076.

Anonymous. 1985. Sales literature. Mametora Agric Machinery Co, Limited, 4 - 33

Roppongi 3 - Chome Minato - ku, Tokyo 106, Japan.

Anonymous. 1985. Nitten paper pots. Sales literature. Mfg. By Nippon Beet Sugar Mfg.
Co, Ltd, No.6, 2 - Chome, Kyobashi, Chuo - Ku, Tokyo, Japan.

Anonymous. 1988. Mechanical transplanter Co, Holland, Michigan, U.S.A 49423.

Anonymous. 2002. Design and development of vegetable transplanter. Biannual report

(2000 - 01). AICRP on FIM (Ludhina Centre) presented during XXIV workshop
held at TNAU Coimbatore (India) during April 18 - 21.

Anonymous. 2004. Report of the all India coordinated research project on farm implements
and machinery. Central Institute of Agricultural Engineering, Bhopal, India.

Anonymous. 2004. Report of the all India coordinated research project on farm
implements and machinery. Department of Farm Machinery and Power, Punjab
Agricultural University, Ludhiana, India.

Anonymous. 2004. Report of the all India coordinated research project on farm
implements and machinery. Department of Farm Machinery and Power, Tamil
Nadu Agricultural University, Coimbatore, India.

Anonymous. 2006. Agricultural Engineering and Technology. In DARE/ICAR Annual

Report 2005 - 06. New Delhi, Department of Agricultural Research and Education,
Ministry of Agriculture, Government of India.

Anonymous. 2008. The New Organic Grower. Chelsea Green Publishing.

Anonymous. 2011. Horticulture Data Base. National Horticulture Board, Department of

Agriculture & Cooperation, Ministry of Agriculture, Govt. of India.

Armstrong, E.C., and W.A. Hanacek. 1984. Automatic soil plug loader, U.S. Patent
No.4,443,151.
Arenas, M., C.S. Vavrina, J.A. Cornell, E.A. Hanlon., and G.J. Hochmuth. 2002. Coir as
an alternative to peat in media for tomato plant production. Hort. Science. 37(2):
309 - 312.

Atiyeh, R.M., N. Arancon, C.A. Edwards and., J.D. Metzger. 2000. Influence of earth
worm processed pig manure on the growth and yield of greenhouse tomatoes.
Biores. Technol. 75(3): 175 - 180.

Boa,W. 1978. The development of automatic transplanters (machinery for block handling
of vegetable plants), Acta Horticulturae. 72: 159 - 169.

Boa,W. 1984. The design and performance of an automatic transplanter for field
vegetables. Journal of Agricultural Engineering Research.30: 123 - 130.

Brewer, H.L. 1978. Automatic transplanter system for field crops. ASAE Paper No.78-
1011, ASAE, St. Joseph, MI 49085.

Branch, G., 1985. A commercially successful automatic transplanter presented. Presented at

54 th Annual Farm machinery Conference, University of California, Davis Jan.8, 3p

Branch, G., 1986. A commercially successful automatic transplanter. ASAE Paper No.86-
1020. ASAE, St. Joseph, MI.

Brewer, H.L. 1994. Conceptual modelling automated seedling transfer from growing
trays to shipping modules. Transactions of the ASAE. 37 (4): 1043-1051.

Brewer, H.L. 1995. Performance of an experimental automatic seedling transplanter with

static cassettes. Journal of vegetable crop production. vol. 1 (1): 37 - 52.

Brewer, H.L. 1997. Increasing planting rates of gravity - fed field seedling transplanters.
Journal of vegetable crop production. vol. 3 (1): 3 - 19.

Bryant, E., S.R. Krogman., and E.J. Mc Ardle. 1984. Plant transfer mechanism. U.S.
Patent. No.4,440,101.

Burgoyne, S. 1987. Comparision of transplanting techniques for processing tomatoes.

Unpublished report to the Ontario Tomato Seedling Grower’s Marketing Board,
Leamington, Ont.
Cayton, D.W., E.C. Armstrong., and A.B. MacKenzie. 1981. Dibble tube soil plug
planter. U.S. Patent No.4,294,179.

Chandra, K.L. 2001. Hand book of Horticulture. Directorate of information and

publication of horticulture, ICAR, New Delhi. pp: 466 - 467.

Chaudhuri, D., V.V. Singh., and A.K. Dubey. 1999. Final report, on Survey of Existing
Status of Seedling/Planting Methods and Determination of Mechanization Needs
for Planting of Important vegetable Crops in India, Central Institute of Agricultural
engineering, Bhopal.

Chaudhuri, D., V.V. Singh., and A.K. Dubey. 2002. Refinement and adoption of
mechanical vegetable transplanters for Indian conditions. Agricultural Engineering
Today. 26 (5 - 6): 11- 20.

Cheong, J., J. Choe, W. Lee., and B. Kim. 1992. Growing mixes and optimum age for
mechanized transplanting of Chinese cabbage. Acta Horticulturae. 319: 437- 440.

Choi, C.W., D.C. Kim., and K.U. Kim. 2001. Development of a pick-up device for plug
seedlings. Journal of Korean Society for Agricultural Machinery. 26(5): 415 - 422.

Choi, W.C., D.C. Kim, I.H. Ryu., and K.U. Kim. 2002. Development of a seedling pick-
up device for vegetable transplanters. Transactions of ASAE. 45(1):13-19.

Chow, J.B., J.W. Wang., and A.L. Myers. 1980. Hand - fed lettuce seedling block
transplanter. Transactions of the ASAE. 23 (5): 1117 - 1120.

Colla, G., Y. Rouphael, G. Possanzini, M. Cardarelli, O.Temperini, F. Saccardo,

F. Pierandrei., and E. Rea. 2007. Coconut coir as a potting media for organic
lettuce transplant production. ISHS. Acta Horticulturae. 747.

Cooley, J. 1991. Introducing plant way. Sales literature. Spring field Nursery, Pick Hill,
Waltham Abbe, Essex EN93LE, England.

Craciun, V., and O. Balan. 2005. Technological design of a new transplanting machine
for seedlings. J. Central Eur. Agric. 7(1):164.

DeGroot, P.F. 1989. Plant transfer mechanism. U.S.Patent No.4,869,637.

Dooley, J.H. 1982. Transplanter for forest nurseries. ASAE Paper No.82-1074, ASAE, st.
Joseph, MI49085,10pp.

Faulring, F.W. 1993. Personal communication on the renaldo automatic transplanter.

4841 Genesee Road, N. Collins, NY 14111.

Feldman, M. 1988. Castle and Cooke transplanter tomato transplanting trial. Engineering
& Statistical Research Centre, Research branch, Agriculture Canada. Ottawa,
Ontario KIAOC6. pp: 37

Feng, D., W. Geng., and D. Zunyuan. 2000. Study on block seedling transplanter with
belt feeding mechanism. Transactions of Chinese Society of Agricultural.
Machinery. 31(2):42 - 45.

Ferminger, T.A. 1953. Firminger’s manual of gardening for India. Thacker, Sprink & Co.
Pvt. Ltd., Kolkata, India.

Garg , I.K., and A. Dixit. 2002. Design, development and evaluation of vegetable
transplanter. Paper presented during 24th workshop of All India coordinated
Research Scheme on Farm Implements and Machinery (ICAR) held at TNAU,
Coimbatore on 18-21 April 2002.

Greer, L., and K.L. Adam. 2005. Plug and transplant production for organic systems.
ATTRA www.attra.ncat.org

Ghai, T. R., and D. Arora. 2007. Horticulture vegetable science (vegetables, tubers and
spice crops), in tropical and subtropical vegetables. New Delhi, India: National
Science Digital Library, National Institute of Science Communication and
Information Resources, Council of Scientific and Industrial Research.

Handreck, K.A., and N.D. Black. 1994. Growing Media for ornamental plants and Turf.
UNSWP.

Hergert, G.B. 1988. Testing of the Lanen transplanter with tomato transplants.
Engineering & Statistical Research Centre, Research branch, Agriculture Canada.
Ottawa, Ontario KIAOC6.
Hergert, G. B. 1988. Testing of the Regero HD B4T transplanter with tomato transplants.
Engineering & Statistical Research Centre, Research branch, Agriculture Canada.
Ottawa, Ontario KIAOC6.

Hergert, G.B., M.Feldman, E.J. Tomecek, M.A. Pelletier, A. Liptay., and F.D. Sumsion.
1988. Investigation of transplanting methods for processing tomatoes. Engineering
& Statistical Research Centre, Research branch, Agriculture Canada. Ottawa,
Ontario KIAOC6.

Huang, B.K., and W.E. Splinter. 1968. Development of an automatic transplanter.

Transactions of the ASAE. 11 (2): 191-194,197.

Huang, B.K. 1973. Seedling growing and handling device. U.S patent No.3,712,252

Huang, B.K., and F. Ai. 1992. Air pruned transplant production system for fully
automated transplanting. Acta Horticulturae. Transplant Production Systems 319:
523 - 527.

Huber, E. 1991. Plant seedling holding and dispensing mechanism. United States Patent
application 5024172, 18 June.

Jeong, B.R. 1992. The use of plug transplants in Korea. Division of Plant Resource and
The environment, College of Agriculture, Gyeongsang National University, Chinju,
Korea. PP. 1 - 9

Kim, H.J., S.H. Park., and T.Y. Kwak. 2001. Development of an automatic transplanter
for cabbage cultivation. Korea Automatic Dynamic Analysis of Mechanical
Systems Conf., Seoul, Korea, 8 - 9 Nov. 2001.

Kohli, S.S., and V.P. Agrawal. 1993. Analysis of a four-bar linkage for a proposed
transplanter. Transactions of the ASAE. 32(2): 317 - 320.

Kuepper, G., and K. Everett. 2004. Potting mixes for certified Organic Production.
Horticulture technical Note. ATTRA www.attra.ncat.org

Kumar, G.V.P., and H. Raheman. 2008. Vegetable transplanters for use in developing
countries: a review. International Journal of Vegetable Science. 14(3): 232 - 255.
Kumar, G.N.M., P.F. Larsen., and K.A. Schekel. 2009. Propagating plants from seed. A
pacific north - west extension publication. PNW0170:1 - 20

Kumar, G.V.P., and H. Raheman. 2010. Volume of vermicompost based potting mix for
vegetable transplants determined using fuzzy biomass growth index. International
Journal of Vegetable Science. 16(4): 335 - 350.

Kumar, G.V.P., and H. Raheman. 2011. Development of a walk-behind type hand tractor
powered vegetable transplanter for paper pot seedlings. Bio systems engineering.
110:189 - 197.

Labowsky, H.J. 2001. Vegetable transplanters. Land Technik. 56 (SH1): 218 - 221.

Lawrence, M.J., D.R. Buckmaster., and W.J. Lamont. 2007. A pneumatic dibbling
machine for plastic mulch. Applied Engineering in Agriculture. 23(4): 419 - 424.

Lee, Y.M., J. K. Wang., and B.A. Kratky. 1982. An improved seedling block
transplanter. Transactions of the ASAE. 313 - 315.

Lloyd Arnold, R. 1990. Plug transplanting automation for processing tomatoes.

Development of transplanting automation technology and volume handling systems
for processing tomato plug plants in Ontario. Ontario tomato seedling growers
marketing board.

Luan, J.M., C. Chi-Liang., and J.S. Ju. 2006. A feeding mechanism of vegetable seedling
transplanter. Chinese Elect. Period. Ser. 14(42): 35 - 42.

Manes, G.S., A.K. Dixit, Surendra Singh, A. Sharda., and K. Singh. 2010. Development
and evaluation of tractor operated vegetable transplanter. Agricultural
Mechanization in Asia, Africa, Latin America. 41(3): 89 - 92.

Margolin, A., Y. Alper., and A. Eshed. 1986. Development of a semi-automatic

transplanter. Acta Horticulturae. 187: 158.

Marr, C.W., and M. Jirak. 1990. Holding tomato transplants in plug trays. Hort. Science.
25(2): 173 - 176.

Marr, C.W. 1994. Commercial vegetable production. Kansas State University,

Manhattan, Kan.
Menzies, N.W., and R.L. Aitken. 1996. Evaluation of fly ash as a component of potting
substrates. Sci. Hort. 67(1-2): 87 - 99.

Moden W.L., and V.L. Hauser. 1982. An automatic fed bandoleer transplanter.
Transactions of the ASAE. 864 - 867.

Munilla, R.B. 1984. A high speed transplanter with extended acceptance and deposition
zones: Design and development. Master of engineering Thesis. Univ. Fla.,
Gainesville.

Munilla, R.D., and L.N. Shaw. 1987. A high speed dibbling transplanter. Transactions of
the ASAE. 30(4): 904 - 908.

Nambu, T., and M. Tanimura. 1992. Development of automatic transplanter using chain
pot for vegetable crops. Acta Horticulturae. Transplant production systems. 319:
541 - 545.

Narang, M.K., I.S. Dhaliwal., and G.S. Manes. 2011. Development and evaluation of a
two row revolving magazine type vegetable transplanter. Journal of Agricultural
engineering. 48(3): 1-7.

Orzolek, M.D. 1996. Stand establishment in plasticulture systems. Hort Technology. 6(3):
181 - 185.

Osamu Sakaue. 1992. Development of automated seedling production and transplanting system
using robotics. Acta Horticulturae. Transplant production systems. 319: 557 - 562.

Parish, R.L. 2005. Current developments in seeders and planters for vegetable

crops. Hort Technology. 15(2):1 - 6.

Paul, L.C., and J.D. Metzger. 2005. Impact of vermicompost on vegetable transplant
quality. Hort Science. 40(7): 2020 - 2023.

Pelletier and G.B.Hergert. 1988. Testing of the Regero HD B4T transplanter with tomato
transplants. Engineering& Statistical Research Centre, Research branch,
Agriculture Canada. Ottawa, Ontario KIAOC6.

Penley, P.A. 1981. High speed transplanter. U.S.Patent No.4,289,080.

Press, R. 2001. Organic gardening. EBSCO Publishing Firm, Ipswich, Mass.

Pugh, C.R. 1983. Total labour input per acre. Unpublished paper, Dept of Economics and
Business. N.C.state Univ., Raleigh. N.C. 27695.

Rotty, R. 1960. Methods and machines used in North American nurseries.Unasylva14(2).

http://www.fao.org.

Rumsey, M.D. 1987. Delivery system for an electro - hydraulic vegetable seedling block
transplanter. M.S.Thesis, University of Hawaii at Manoa.

Rumsey, M.D., Y.N. Ting., and J.K. Wang. 1989. An improved electro-hydraulic vegetable
seedling block transplanter. Applied Engineering in Agriculture. 5(4): 485 - 488.

Ryu, K.H., G. Kim., and J.S. Han. 2001. Development of a robotic transplanter for
bedding plants. Journal of Agricultural Engineering Research. 78(2): 141- 146.

Sachan, C.P., S.K.Singh., and B. Singh. 2010. Growing vegetables with quality seed is
remunerative. Indian Horticulture. March - April 2010: 46 - 49.

Satpathy, S.K., and I.K. Garg. 2008. Effect of selected parameters on the performance of
a semi - automatic vegetable transplanter. Agricultural Mechanization in Asia,
Africa, Latin America. 39(2): 47 - 51.

Schmilewski, G. 2008. The role of peat in assuring the quality of growing media. Mires
and Peat. http://www.mires-and-peat.net/, ISSN 1819-754X.

Schofield, A . 2007. Plant raising - a practical approach for growers. The Organic Grower
.No 2, 22-24. The Journal of the Organic Growers Alliance.

Shaw, L.N. 1986. Towards automatic transplanting. Proc.Fla.State Hort. Soc.99 : 300 - 302.

Shaw, L.N. 1988. Spatial transplanter mechanism. U.S.Patent No.4,788,920.

Shaw, L.N. 1997. Automatic transplanter for vegetables. Proc.Fla.State Hort. Soc.110:
262 - 263.

Shaw, L.N. 1999. Removing and handling modular vegetable seedlings from nursery
trays. Proc. Fla. State Hort. Soc.120: 153 - 155.
Singh, B., H.L. Yadav, M. Kumar., and N.P.S. Sirohi. 2007. Effect of plastic plug tray
cell size and shape on quality of soilless media grown tomato seedling. Acta
Horticulturae. 742: 57 - 60.

Singh, R.S., and V.V. Singh. 2009. Economic evaluation and mechanization gaps of vegetables
cultivation in Madya Pradesh. Agricultural Engineering Today. 33(2): 32 - 36.

Singh, B., B.S. Tomar., and M. Hasan. 2010. Plug tray nursery raising technology for off
season cucurbits cultivation. Acta Horticulturae. 871: 279 - 282.

Sena, G.A. 2006. Adjustable tray size automatic seedling planting apparatus. United
States Patent application 7036440, 2 May, 2006.

Srivastava, A.K., C.E. Goering, R.P. Rohrbach., and D.R. Buckmaster. 2006.
Engineering Principles of agricultural machines. American Society of Agriculture
and Biological Engineering. St. Joseph, Mich.

Subler, S., C.A. Edwards., and J.D. Metzger. 1998. Comparing vermicomposts and
composts. Biocycle. 39(7): 63 - 66.

Suggs, C.W., T.N. Thomas, D.L. Eddington, H.B. Peel, T.R. Seaboch., and J.W Gore.
1987. Self - feeding transplanter for tobacco and vegetable crops. Applied
Engineering in Agriculture. 3(2): 148 - 152.

Suggs, C.W., D.L. Eddington, H.B. Peel., and T.R.Seaboch. 1989. Self -feeding
transplanters. Proc.11 th CIGR Congress on Agri. Engr., 3:1715-1722, Dublin, Ireland.

Suggs, C.W, B.M. Lineberger., and S.C. Mohapatra. 1992. Automatic feeding
transplanter. Transplant production system. Acta Horticulturae. 319: 511- 516.

Tesch , S.M., and B.L. Bierman. 1993. Seedling array transplanter. United States Patent
application 5215550, 10 July, 1993.

Todd, G.K. 1972. Seedling flat. U.S.Patent No. 3,667,15.

Tsuga, K. 2000. Development of fully automatic vegetable transplanter. Japan. Agric.

Res. Q. 34(1): 21- 28.

Williams, G.A. 1990. Seedling planting machine. U.S. Patent No. 4,970,972.

Williames, G.A. 1997. Apparatus and method for transferring seedlings from plant

trays. United States Patent application. 5644999, 8 July, 1997.

Trends in Biosciences 7(3): 407-410, 2014
Development of a Gravity Fed Automatic Vegetable Transplanter with Walking
Beam Mechanism
S. SIVAKUMAR AND C. DIVAKER DURAIRAJ
Department of Farm Machinery, 2AEC&RI,Tamilnadu Agricultural University,
email : sivakumar95027@gmail.com
ABSTRACT
A gravity fed automatic vegetable transplanter with walking
beam mechanism for feeding the plug tray was developed and
evaluated in the laboratory. The main objective of this work is
to make the seedling drop by gravity without causing damage
to the seedling. To make the feeding pro - tray simple, an open
bottom type pro tray seedling is used. To move the tray
continuously, a 50 rpm gear motor with walking beam type
moving mechanism was developed and tested. The
conventionally practiced growing medium with 20, 30 and 40
days old seedling is used for conducting experiments. Under
laboratory conditions it was found that it was feasible to drop
28 plants min-1.
Key words Transplanter bean mechanism, automatic
vegetable transplanter
India has emerged as a leading horticultural country
of the world with a total annual production of 240 million
tonnes of horticultural crops during 2010 - 11. Developing
countries contribute 72 per cent of the total vegetable
production in the world. India is the second largest producer
of vegetables after China. Production of vegetables in India
stands at 14.0 per cent of the world production in 2010 -
11. India produces nearly 60 leafy, fruity and tuber varieties
of vegetables. The area under vegetable is around 8.50
million ha and production 146.55 million tones (Horticulture
data base 2011). At present the transplanting of vegetable
seedling is carried out manually all over the country. The
vegetable nursery is grown in the field and later transplanted
to the main field by manually transferring them but their
survivable is usually not good. Nowadays, the vegetable
growers have moved on to the pro - tray seedling, since
the pro - tray seedling has good reestablishment capability
after transplanting them manually or mechanically.
Some minor dimensional changes may have to be
made to the trays to make mechanical handling easier
(Shaw, 1986). The tray grown plants cost more but their
survival is better and transplanting shock is much less.
The tray grown seedlings have stirred up interest in
developing fully automatic transplanters, since the seedlings
are in an orderly array that might be handled mechanically
(Shaw, 1999). The difficult problem is how to remove the
seedlings from the tray.
The imported vegetable transplanters are not suitable
for Indian soil condition. Moreover the existing mechanism
of plant removal in these machines are highly complex,
hence this attempt is to simplify the plant removal
mechanism for using appropriate methodology. Automatic
dropping of seedling by gravity with continuous movement
of pro - tray was hence contemplated.
MATERIALS AND METHODS
The seedling instead of being taken from top is planned
to be dropped through the bottom. The plug - tray, which
is open at the bottom, is not commercially available. So
pro - tray cell bottoms were manually cut and used on trial
basis. Instead of using the whole pro - trays, pro - tray
was segmented into seven linear strips and used, so that
each strip with seedling can be fed to the machine rather
than the whole tray.
The pro - tray strips were cut manually to uniform
size and the bottom was removed manually for which a
dye was made and placed on top of the pro - tray and knife
was used to cut the bottom of the pro - tray. Figure 1
shows the details of the open bottom type pro - tray
segments (Tsuga, 2000). It was felt that if this method
proved successful, the seven segments could be made
attachable to each other to grow the seedling in a whole
unit, after which they could be separated and fed into the
designed machine. The important difference here is that
the pro - tray segment was placed upside down, media
filled and seeds sown in bottom-up position. When the
seedlings are fully grown (Fig. 2) the taper available is the
cell will allow the seedling to slide down easily by gravity.
A feeding chute for the machine was fabricated which
would allow the segmented pro - trays into the machine
from waist height. The experimental feeding chute was
developed to test the process by feeding the pro - tray
strips towards testing whether gravity ejection is really
feasible. The details of the chute are shown in figure 3. A
transfer plate was also developed for transferring the pro -
tray segments with seedlings into the feeding chute (Fig.3).
While transferring the seedlings to the transfer plate the
seedlings are loosened slightly by gravity from the pro -
tray due to the two round rods provided on the inner edge
of the transfer plate and the clearance created thus. The
408 Trends in Biosciences 7 (3), 2014
Fig. 1. Open bottom type pro – tray
transfer plate was then inserted onto the top of the feeding
chute and the seedling tray pushed down by hand, so that
it slides downward until it reaches the bottom of the feeding
chute. At the bottom end of the parabolic sliding feeding
chute, a matching hole is provided. When the pro - tray
cell bottoms are synchronized in position with the bottom
hole of the feeding chute at the end, the seedling could fall
down by gravity to the ground. To prevent tipping of the
seedling to side while moving down the chute, ribs were
provided on both sides of the feeding chute with a height
of 1.5 cm.
A 12 V, DC, 50 rpm centre shaft gear motor was
used to provide the movement to the seedling segment.
The circuit that allowed the push rod to push the tray cells
one by one is shown in figure 5. A limit switch was provided
which stops the movement after, the tray segment moves
through one cell distance. When the hole is aligned to the
seedling bottom, the seedling is free to fall through it. A
push button was connected in parallel to override the limit
Fig. 2. Growing of seedling in open bottom pro - tray
segments
Fig. 3. Test rig for seedling ejection by gravity
switch when next movement was required. Thus the tray
segment carrying the seedlings was pushed one by one to
the hole for making them to fall through.
The motor is connected through the NC terminals of
the limit switch to the battery. So when the finger activates
the limit switch, the circuit is opened and the motor is
stopped and is retained in that position. But when the push
button is pressed manually and momentary, the circuit is
completed and motor rotates till the next finger hits the
limit switch and so on.
Walking beam mechanism using double crank
with belt drive:
This is basically a double crank mechanism with the
driven and driving radius being equal. The geared motor
drives one of the crank which is a toothed pulley with
1mm pitch. On the top of the pulley a crank pin was setup
which connects to the driven crank through a connecting
rod of flat mild steel. The driven crank is also a toothed
Fig. 4. Walking beam mechanism with belt drive
 SIVAKUMAR AND DURAIRAJ, Development of a Gravity Fed Automatic Vegetable Transplanter with Walking Beam Mechanism 409
Fig. 5. Circuit diagram for imparting movement to the pro - tray segment
pulley of 1 mm pitch and carries a crank pin on its top to
connect the connecting rod. The two pulleys are meshed
with a 1 mm pitch toothed belt of 8mm width and 200 mm
length. When the motor rotates, the crank mounted on its
shaft drives driven crank pulley at the same angular
displacement. The connecting rod hence moves in
synchronization with the pulleys and the crank always
moves parallel to the line connecting the centres of the
double crank. A push rod mounted perpendicularly to the
connecting rod moves with the connecting rod with its
end making a circular motion. The crank radius being 1.8
mm, the push rod moves in a circle of 3.6 mm diameter.
The push rod moves into the gap between the cells at one
top dead centre of the mechanism, moves the cell through
3.6 mm axially along the test rig pathway and releases out
of the gap at the other top dead centre. This provides a
perfect axial (working beams) motion to the tray cells. The
Fig.6. Effect of gravity falling with open bottom type pro
- tray
motor was driven by a 12V battery. This type mechanism
did not damage the cells and was perfectly aligning the
cells to the hole. The walking beam mechanism is shown
in figure 4.
RESULTS AND DISCUSSION
The seedlings were grown in open bottom type pro -
tray with growing medium of coir pith + vermicompost
(4:1) (media 1), coir pith + vermicompost (4:1) + paper
cone (media 2), soil + vermicompost (4:1) (media 3), and
soil + vermicompost (4:1) + paper cone (media 4), and
tested on the developed rig. The agronomical parameters
like plant height, stem diameter, no of leaves and root weight
were measured to find any influence of the media on the
growth when grown in open bottom type pro - tray. It was
found that no significant difference existed between them
(Fig 6). It was observed that the (media 4) soil +
vermicompost (4:1) + paper cone seedlings falls down
successfully due to the weight of the seedling when
compare to other mediums. Moreover the paper cone
prevents media sticking on to the pro - tray hence all the
seedlings were able to fall down successfully without any
damage to the root ball.
The mechanism was smoothly moving the pro - tray
segment. Since the push rod was pushing the cells
tangentially, thus moving axially along the test rig, they do
not cause any damage to the plastic wall of the cells, 28
plants were falling out successfully per minute. The
following problems were observed in the experiment.
1. Though the test rig had a parabolic path way to slide,
the segment carrying the seedling was quite heavy
and slides too fast to the bottom and hit the feeder
mechanism with a force. The subsequent feeds caused
than to override the trays fed previously. This
410 Trends in Biosciences 7 (3), 2014
necessitated to add one more intermediate drive of
feed the trays slowly, which would turn out to be
complex and was not done.
2. The segmented tray could not be manually sliced with
precision and took some lateral swing while travelling
down the rig carrying misalignment.
3. Since the seedling has to fall through the cell, older
seedlings which have more vegetative growth get
caught on the top of the cell while falling through.
4. Though the paper cone improved the scouring to some
extent, watering of tray wets the paper and makes it
ineffective. Altogether the ejection by gravity was felt
unsatisfactory.
ACKNOWLEDGEMENT
The research work was supported by the Council of
Scientific and Industrial Research, Human Resource
Development Group (Extra Mural Research Division), New
Delhi under research fellowship program. The authors are
grateful to Department of Farm power and Machinery,
Agricultural Engineering College and Research Institute,
Tamil Nadu Agricultural University for providing facility to
carry out the research.
LITERATURE CITED
Horticulture data base 2011. National Horticulture Board, Department
of Agriculture & Cooperation, Ministry of Agriculture, Govt. of
India.
Shaw, L.N.1986. Towards automatic transplanting. Proc.Fla.State
Hort. Soc. 99: 300-302.
Shaw. L.N.1999. Removing and handling modular vegetable seedlings
from nursery trays. Proc. Fla. State Hort. Soc. 120: 153-155.
Tsuga, K. 2000. Development of fully automatic vegetable
transplanter. Japan. Agric. Res. Q. 34(1):21–28.
Recieved on 12-01-2014 Accepted on 02-02-2014
Trends in Biosciences 7(4): 621-624, 2014
Development of A Seedling Ejection Mechanism for Pro - tray Seedling
S. SIVAKUMAR1 AND C. DIVAKER DURAIRAJ2
1
Department of Farm Machinery, 2Agricultural Engineering College & Research Institute, Tamilnadu
Agricultural University, Coimbatore- 641 003,Tamil Nadu
email : sivakumar95027@gmail.com
ABSTRACT
A novel seedling pick - up mechanism was developed and its
performance was evaluated at the laboratory. The pick - up
mechanism extracts seedlings from a 98 cell pro - tray and
transfer them to place into the soil. The pick - up mechanism
consisted of a coupler extension (claw), driven link, driving
crank, and coupler. The pick - up mechanism is a four bar
mechanism modified according to our requirement. The
coupler joins the driving link and driven link. When the crank
rotates, the claw enters into the pro - tray cell from the top and
tease out a seedling from its cell. The coupler path was so
designed to do this job without damaging the root ball and the
plant. The pick - up mechanism was tested with different
medium and depth of penetration. When tried on 20 days old
seedlings, The seedling pick - up mechanism extracted 15 to
20 seedlings per minute and the success ratio was 80 per
cent.
Key words Pro-tray, Tray holder, Pick - up claw, Four bar
mechanism
One of the important components of the vegetable
transplanter is the seedling extraction mechanism, which
extracts the plug seedling from the pro - tray cell and
transfers them to the transplanting device, which places
them into the soil. Yanmar Agricultural Equipment Co. and
Kubota Cooperation, leading agricultural machinery
manufacturers in Japan, have developed two of the most
common seedling pick - up devices for the vegetable
transplanters widely used in Korea and Japan. The Yanmar
- type moves the pick- up pins toward the lower part of
the cell surface and extracts the seedling from the cell while
moving along a path in an open, counter - clockwise loop.
The Kubota - type generates a crossed path when picking
up seedlings by using a more sophisticated mechanism
comprising of a slider, cam, and links. Although these two
pick - up devices perform well, their structural complexity
have made them difficult to use for various types of
vegetable transplanters. In addition, these pick - up devices
are not economically feasible for indigenously made
vegetable transplanters because of their high manufacturing
costs (Choi, 2001).
The demand for mechanization of vegetable
production in India has increased every year, so attempts
are being made to develop an automatic transplanter for
vegetables. Seedling pick - up mechanisms that are simple,
accurate and economically feasible are to be developed for
vegetable transplanters in India. This study was one of
these attempts. The articles authored by Brewer, 1994, Choi,
2001 and Choi, 2002 were reviewed to develop a concept
of seedling ejection device that will satisfy our requirement.
Not much of literature is available in this context. The
objective of this study was to develop a seedling pick - up
mechanism for vegetable transplanters suitable to Indian
conditions and evaluate its work performance in a
laboratory.
MATERIALS AND METHODS
Development of seedling ejection mechanism:
To obtain the best performance of the pick - up
mechanism, it needs to satisfy the following requirements.
1. The pick - up claw enters into the pro tray cell with
the depth of penetration not to be more than half the
depth of the pro - tray cell
2. The pick - up claw should release the seedling
immediately after coming out from the tray cell.
3. The mechanism should be simple and easy to fabricate
4. Extracting and discharging of seedling should be done
at predetermined points on the locus of the claw.
5. The stem and leaves of the seedling should not be
damaged
6. The root ball of the seedling should not be damaged
during extraction of the plug
Experiments on seedling ejection by clawing out
from the top of the pro - tray cell:
The developed pick - up mechanism claws out the
seedling plug from the top of the tray using an appropriate
mechanism. It was felt that the four bar mechanism would
be appropriate to try this method by parking the seedling
trays on its side and the mechanism opposite to the cell. A
four mechanism was designed to extract the seedling from
the tray cell and release it to fall free to the ground. The tip
of the claw as an extension of the coupler bar was supposed
to extract the plug seedling entering into the top of the pro
- tray cell and tease it out, by moving in a predefined path.
622 Trends in Biosciences 7 (4), 2014

Fig. 1. The finalized coupler curve of the mechanism

But care should be taken not to hit the seedling stem when Experimental setup for assessing extraction
it is done. A freeware by the name “four bar” was used to performance of finalized mechanism:
synthesise the coupler curve of the designed mechanism.
A tray holder was (Fig 2) developed wherein the pro
A path was first assumed to do the job and the coupler
- tray with seedling could be placed horizontally. Three
curve analysis using the freeware was done to optimize
pair of six mm rod with height of 30 cm was mounted on
the dimensions of the four bar mechanism appropriately to
the holder platform vertically at equal distance with 2 mm
generate the required path of the claw. Figure 1 shows the
coupler curve finalized for the mechanism.

Fig. 2. Tray holder

Fig. 3. Seedling pick - up mechanism
 SIVAKUMAR AND DURAIRAJ, Development of A Seedling Ejection Mechanism for Pro - tray Seedling
Fig. 5. Success ratio by penetration depth of claw
gap between the rods for inserting the pro - tray from
above. The tray holder was made to slide horizontally on a
rectangular bar and the whole assembly was fixed on a flat
wooden plank. A handle was fitted at end of the tray holder
for moving the pro- tray forward or rearward so that the
pro - tray could be moved. The pro - tray along with the
seedlings could hence be moved and positioned against the
ejecting mechanism, so that ejection performance could
be assessed against each seedling in a row. For assessing
the bottom or top row, the holder platform was raised up
on a set of wooden blocks to the required height. The four
bar mechanism was mounted on a vertical stand with a
handle provided for hand cranking (Fig 3). The handle when
rotated in the appropriate direction made the pick - up pin
of the claw enter the top side of the pro - tray to a depth of
3 cm, teased and ejected out the seedling. The pick - up
claw was expected not to break up the root ball during the
Fig. 6. Success ratio by claw shape
623
Fig. 7. Success ratio by rotation speed
removal. In order to avoid root ball damage, the tips of the
claw were flattened and the pro - tray seedlings removal
was successful.
RESULTS AND DISCUSSION
The 98 cell pro tray with plug seedlings was mounted
vertically and the seedling pick - up mechanism was installed
in front of the tray holder. The distance between the seedling
tray and pick up mechanism was manually adjusted. Figure
4 shows the pick - up claws extracting the seedling one by
one from the tray. The vegetable seedlings used for
evaluation were 20 days old seedlings as grown in three
media namely coir pith alone (M1) , coir pith + vermin
compost 4:1 (M2) and coir pith + soil 1:1 (M3) (by volume).
The seedlings were watered a day before conducting
the performance test in the laboratory. The performance
Fig. 8. Success ratio by age of seedling
624 Trends in Biosciences 7 (4), 2014
Fig. 9. Success ratio by growth medium
test was conducted under 4 sets of operational conditions
such as depth of operation, age of seedling, speed of rotation
and growth medium. From the figure 5 it is observed that
the pick - up claw penetrated into the root media to a depth
of 35 mm, which was found satisfactory with the success
ratio of 82 per cent.
When the crank was rotated by hand at appropriate
speeds of 25 and 50 rpm, the success of seedling extraction
was 67 and 81 per cent respectively, because the higher
the rotational speed it had caused slippage of the seedling
in the tray (Fig. 7). Most of the successful extractions
were made on 20 days old seedlings, while using the 30
and 40 days old seedlings the leaves get on tangled, which
prevents free falling of the seedling. The growth medium
does not affect the performance of the seedling (Fig.9).
A seedling ejection mechanism for extracting the plugs
out of the pro - tray was developed and evaluated in a
laboratory. The pick - up claw extracts the seedling from
the tray cell. This is a four bar mechanism comprising of a
driven link, driving link and pick up claws. This mechanism
was tested under various operational conditions such as
seedling age, penetration depth, growth medium and speed
of rotation. The pick - up device extracted 15 to 20 seedlings
per minute and the success ratio was 80 per cent, while
tested on 20 days old seedling. The seedling removal was
successful but the root ball being fragile often broke up. In
future, adding some binding material with growth medium
could be used to improve the root ball integration.
ACKNOWLEDGEMENT
The research work was supported by the Council of
Scientific and Industrial Research, Human Resource
Development Group (Extra Mural Research Division), New
Delhi under research fellowship program. The authors are
grateful to Department of Farm power and Machinery,
Agricultural Engineering College and Research Institute,
Tamil Nadu Agricultural University for providing facility to
carry out the research.
LITERATURE CITED
Brewer, H.L. 1994. Conceptual modelling automated seedling transfer
from growing trays to shipping modules. Transactions of the
ASAE 37 (4): 1043-1051.
Choi. C.W., Kim, D.C. and Kim, K.U. 2001. Development of a pick-
up device for plug seedlings. Journal of Korean Society for
Agricultural Machinery 26(5): 415 - 422.
Choi, W.C., Kim, D.C., Ryu, I.H. and Kim, K.U. 2002. Development
of a seedling pick-up device for vegetable transplanters. Trans.
Amer. Soc. Agric. Eng. 45 (1):13-19.
Recieved on 11-02-2014 Accepted on 25-02-2014