Everglades Station Mimeo Report EES66-8 February 1966
A MECHANICAL HARVESTER
FOR
FRESH MARKET CELERY
BY
James F. Beeman
Asst. Professor and Asst. Agricultural Engineer
Department of Agricultural Engineering
University of Florida
William W. Deen, Jr.
Asst. in Agricultural Engineering
Everglades Experiment Station
Belle Glade, Florida
lawrence H. Halsey
Asst. Horticulturist
Department of Vegetable Crops
University of Florida
Gainesville, Florida
For Presentation at the 1966 Southeast Region Meeting
AMERICAN SOCIETY OF AGRICULTURAL ENGINEERS
Jackson, Mississippi
February 7-9
Page 2
A MECHANICAL HARVESTER FOR FRESH Page 2
MARKET CELERY
By
James F. Beeman, William W. Deen, Jr., and
Lawrence H. Halsey
INTRODUCTION
The state of Florida produces about 30 percent of the celery grown in the
United States. In the 1963-64 season over 11,000 acres of fresh market cegry were
grown in the state and represented a farm income of over $20,000,000 (2) -'.
During the 1964-65 harvest season approximately 500 acres of celery were not harvest-
ed. A large portion of this loss was attributed to the unavailability of harvest
labor.
Nearly all the celery grown in Florida is cut by hand and packaged in the
field on mobile packing houses or "mule trains". According to Spurlock (4) an
average harvesting crew in the agricultural Everglades region during the 1964-65
season consisted of approximatel y 69 persons. This harvest crew was utilized as
shown in Table I.
Table 1. Number of Workers and Man-Hours 2er Acre to Harvest and Pack
Celery -=
Function Man-Hours Per Acre Number of Workers
Cut, strip and top 49.5 19.5
Grade and pack 71.1 28.7
Make, set off, close crates 23.4 9.5
Load and drive trucks 16.6 6.2
Other and supervision 14.0 5.3
174.6 69.2
The present system requires the walking hand cutter to bend at the waist, hold
the stalk while severing the root stem at ground level with a knife, stand erect,
strip the undesirable petioles from the stalk, and place it on a waist high conveyor
belt. The stalks are aligned by hand before being topped by a circular saw during
travel on the belt. The stalks then enter the mobile packing house where they are
washed, stripped further if necessary, and packed in crates according to size.
The task of the hand cutter is the most demanding of any performed during the
harvesting-packing operation and is the one for which it is difficult to employ
workers. The normal harvest rate is between four and seven feet per minute, which
requires each hand cutter to cut and strip over 5,000 plants per day. This opera-
tion becomes especially objectionable during inclement weather, during which time
the complete harvesting-packing operation may be halted because of the inability of
the cutter to function.
A machine which would mechanically cut the celery,strip the tops, and elevate
the stalks to a position where they could be further manipulated could help
alleviate some of the growers' harvesting problems. In addition to the labor sav-
ing features, the integration of a mechanical harvester into the complete harvest.-
ing-packing operation would provide an opportunity to make the overall system
more efficient.
/ Numbersin parenthesis refer to references.
Based on average of 811 crates per acre.
Page 3
COMPARISON OF PRESENT AND PROPOSED
HARVESTING-PACKING OPERATIONS
A flow process chart provides one of the best means for analyzing the present
and proposed systems. The basic steps involved in the present system of cutting,
packing, and transporting celery to the precooling facilities are presented in
Figure 1. This system involves 16 distinct hand operations, most of which must be
performed under direct exposure to the environment.
Two alternate systems which could utilize a mechanical harvester will be com-
pared with the present system. These are (1) a central packing house-precooling
system and (2) a modified mule train system. Both of these plans include the use
of self-unloading vehicles in which the celery from the harvester can be transport.-
ed in bulk to the packing facilities.
The flow process chart for the central packing house-precooling system is
shown in Figure 2. This system could utilize automatic weight sizers and crate
closer. By comparing Figure 2 with Figure 1 (the present system) it can be seen
that the use of the central packing house system could reduce the number of hand
operations from 16 to 4, all of which could be performed in the packing house.
Since many producers do not have suitable packing house facilities, but do
have from 10,000 to $30,000 invested in each mule train, the modified mule train
system is proposed. A flow process chart for this system is showrtin Figure 3.
This plan would not be as efficient as the central packing house plan in the re-
duction of hand operations, but it would provide a means for utilizing present
equipment and would reduce the number of hand operations to a total of 11.
OBJECTIVES
The objectives of this study were to develop the functional specifications
for a mechanical celery harvester which would reduce the amount of stoop labor
required to harvest the crop and which could eventually -be integrated into a more
fully mechanized system for harvesting and handling fresh market celery.
The study consisted of two parts which were carried out simultaneously; (1)
the determination of the physical and mechanical properties of the celery plant and
(2) experimentation with various components which could be used on an experimental
harvester.
PHYSICAL PROPERTIES
Considerable knowledge of the nature of the product is very important in the
development of machinery for the manipulation of agricultural products. The
physical properties studied were those which could provide information useful in
the development of the harvesting machinery and of the subsequent bulk handling
operation. These properties included: plant profile, plant weight, coefficient of
static and sliding friction, center of gravity, force to cut root stem, and damage
resistance. All studies were conducted with a variety designated 213 which repre-
sents about 95 percent of the celery grown in Florida. A detailed description of
how most of these physical properties were measured along with the results obtain~l
were presented in a previous publication (1). Values for several of the physical
properties measured are listed in Table 2.
The coefficient of static friction for whole plants on plywood, aluminum,
black iron, and rubberjized be].lt.ing we-r 0.96, 1.49-, 1.(C. pnr 0.75, r-'oectively.
FLOW PROCESS CHART
Present Method For Handling Celery Cutting To Precooler
7)
0
6;
SK~
Figure 1
Crate
Crate bundles stored on vendor's
truck at field edge
Load.bundles on field trucks
To mule train unit
Wait for previous truck to load
Inook truck to mule train and unload
bundles,
Store on second deck of mule train
Crate stack pulled to work1 position
Bundle strings cut
Make up crate, set on delivery slide
Store on delivery slide
Set up for packing on lower level
of mule train
Wait for packer
Position for packing
Page 4
Celery
O Cut stalk, strip, place on
conveyor
STo top cutter
O positioned manually for top
cutter
STop cut by rotary saw
STo washer
r-, Wash with water spray
|') To packers
- i
Sized and sorted
SStripped and packed in crate
SPosition crate for preclosing
To. closing station
0 Clqse lid
j To truck.
SLoaded on truck
`7 Accumulate load on truck
jI Release truck from mule train
K' To precooling plant
Position for unloading
SUnload onto conveyor
STo precooler
-\To
___ ------II~----
Page 5
ALTERNATE SYSTEM ILTUvBER 1
CENTRAL PACKING HOUSE
Self Propelled Field Wagon
Mechanical Harvester
Prepare to drive away from
unloading dock at precooler
To field
To end or center of field
Wait for harvester to reach
position
Position for loading
Crates
Bundles stored on vendor's truck
at packing house
Unload bundles with lift truck
\ Transport bundles to packing area
1 Bundle strings cut
Make up crate set on delivery slide
Store on delivery slide
Wait for packer
Position for packing
gu
C To field once per day
Position for harvesting
T
I
0
.re 2
b
0
--f
Celery
Stalk cut, topped, partially
stripped, loaded in wagon
Accumulate load on wagon
To central packing house
Position for unloading
Unload on to conveyor belt
To strippers
Stripped and sorted
To sizer
Machine size
To packer
Packed
To closing sta+.ion
Machine close
To prY-olAtr
(
)
--I-"--"~-
Fi
6
6?
Wait for packer
Position for packing
Strip and sort
To packers
ISize and pack
To closing station
CGlose crates (mnchine or maniiln)
STo truck
SLoad on truck
Accumulate load on truck
Figure 3
,To Precooler
ALTERNATE SYSTEM NUMBER 2 Page 6
Mule Train Located at the End of the Field
Self Propelled Field Wagon Mechanical Harvester
I-repare to drive away from garage j To field once per day
To field Q Position for harvesting
To end or center of field
Wait for harvester to reach position
Position for loading Celery
Celery
Crates Stalk cut, Partially stripped,
C-: loaded in wagon
Bundles stored on vendor's truck
parked at mule train /Accumulate load on wagon
To top of mule train by small crane me tain a e o
on top of mule train L To mule train at end of field
on top of mule train /
Bundle strings cut Q Position for unloading
Make up crates, set on delivery slide Unload on conveyor belt
Store on metal delivery slide To washer
Set up for packing on lower level Wash with water spray
Page 7
Table 2. Physical Properties
Mean Standard
Measurement Value Deviation Range
Plant Weight 3.47 lbs. 1.00 lbs. 3.32 lbs.
Plant Height 28.17 in. 2.17 in. 7.50 in.
Root Stem Diameter 1.30 in. 0.211 in. 0.72 in.
Center of Gravity
(from base)
Untopped, Unstripped 8.09 in. 1.09 in. 2.50 in.
Topped, Unstripped 6.90 in. 0.33 in. 1.0 in.
Plant Width
(perpendicular to row)
at 4 in. height 23.86 in.
at 8 in. height 24.86 in.
at 12 in. height 25.88 in.
at 16 in. height 24.02 in.
at 20 in. height 22.61 in.
at 24 in. height 20.10 in.
A mean force of 32.2 pounds was required to cut the root stems when the blade
waOs positioned at an angle of 60 degrees with respect to a normal to the line of
travel. The required force increased with decreased blade angles due to less
slicing action. A linear regression analysis on the relationship between the force
to cut and the blade angle yielded
Y = 53.21 0.2556X
where Y is the force in pounds and X is the blade angle in degrees. The correlation
coefficient is 0.38, indicating that the relationship may not be linear. Another
statistical analysis indicated the absence of a simple relationship between force
to cut and mean root diameter or root cross sectional area.
Resistance to drop studies indicated that damage to marketable petioles begins
when unstripped, topped stalks are dropped on their side from a height of approxi-
mately 12 inches. The minimum height at which damage occurred appeared to be a
function of the weight of the stalk. A linear regression analysis of this relation-
ship yielded
Y = 46.75 11.86 X
where Y is the height in inches and X is the weight of the stalk in pounds. The
correlation coefficient is 0.509.
A bulk handling test was conducted to study the feasibility of transporting
mechanically harvested celery in self unloading vehicles to the packing facilities
and to determine the effect of stacking height upon damage.--I Unstripped stalks
were used so that the unmarketable petioles, which would be removed at the packing
facilities, would provide some protection for the marketable petioles.
A self unloading, wooden trailer with dimensions 4 feet deep, 7 feet wide, and
14 feet long was used to transport the celery. The stalks were manually cut and
topped and then carefully stacked in oriented arrangement in the trailer in two,
three, and four foot depths. The loaded trailer was then pulled approximately
seven and one-half miles over fairly rough road. The celery was next evaluated for
damage to the marketable petioles. Evaluations were conducted at the floor to si:
inch level and at every foot level. Four positions were selected along the width
of each level and six stalks were sampled at each of these four positions. Two
replications were conducted. The results are shown in Tables 4 through 10.
3/ Appreciation is acknowleedg to William G. Grizzell, Tndustrial Engine r, USDA-
ARS, Transportation and Facilities Rese-arch TDiAvsion, for his assistance in condlnct-
ing the bulk handOling test,
Table 4.
Celery Petiole Damage
Bulk Handling Celery Test No. 1 -- Depth 2 f-.
Location in
1/
Trailer-
- .oftf quarter
Center
Right Quarter
2S,K /-
K
2K,S 2K,S 2(
N B :B
3K
K,S
K
N
S,K K
K S
K N
K,S K
1/ Trailer was 7 ft. wide. Dimension location of samples were 1-3/4, 3-1/2 and 5-1/4 ft. from the right
trailer wall to the right quarter, center, and leff quarter, respectively.
2/ Code: K--break; B--bruise, S--scuff; ,--no ip ury; C--crack.
Depth
Level
C,K N
2K S
Inches
24
12
6
Wall
Location in
LYf+-auate
Table 5.
Celery Petiole Damage
Bulk Handling Celery Test No. 1 -- Depth 3 ft.
Location in
I /
Trailer-
Left quarter
Center
Right quarter
2K,B N
N K
3K C,S 2K
3K,S B,S 2S
2K,S S
N S
N 2K
K,C,B K
1/ Trailer was 7 ft. wide. Dimension location of samples were 1-3/4, 3-1/2 and 5-1/4
trailer wall to the right quarter, center, and left quarter, respectively.
ft. from the right
2/ Code: K--break; B--bruise; S--scuff; N--no injury; C--crack,
3/ After settling.
Depth
Level
Inches
3/1/
31-
2/
K
24
12
6
Right nuarter -Wall
K,S N
K 2K
K,B 2K
B K
2K
C
N
K,S
N
N
K,S
N
L
Wall
Table 6.
Celery Petiole Damage
Bulk Handling Celery Test No. 1 -- Depth 4 ft.
Location in
1/
Trailer-
Left' quarter
Center
Right Quarter
2/
K
N
K,S
N
N N K,C
K,S 2K,C K
1/ Trailer was 7 ft. wide. Dimension location of samples were 1-3/4, 3-1/2, and
trailer wall to the right quarter, center, and left quarter, respectively.
2/ Code: K--break; B--bruise; S--scuff; N--no injury; C--crack.
5-1/4 ft. from the right
3/ After settling.
Depth
Level
Inches
3/
41-
36
24
12
6
Wall
B,S
N
Table 7.
Celery Petiole Damage'
Bulk Handling Celery Test No. 2 -- Depth 2 ft.
Location in
1/
Trailer-
Left quarter
2/
N-
K
Center
Right quarter
N
K,S
N
N
N
N
1/ Trailer was 7 ft. wide. Dimension
trailer wall to the right quarter,
location of samples were 1-3/4, 3-1/2 and 5-1/4 ft. from the right
center, and left quarter, respectively.
2/ Code: K--break; B--bruise; S--scuff; N--no injury; C--crack.
3/ After settling.
Depth
Level
Inches
23/
20-
Wall
Lnf quarter
Table 8.
Celery Petiole Damage
Bulk Handling Celery Test No. 2 -- Depth 3 ft.
Depth
Level
Left quarter
Inches
3/ :
281.
24
12
6
1/
Location in Trailer-
Center
2/
2K-
K
K
N
2B
N
K
N
N
K,B
S
K,B,S
N
N
Right quarter
N
N
B,K
B
2S
S
B,S
K
1/ Trailer was 7 ft. wide. Dimension location of samples were 1-3/4, 3-1/2 and 5-1/4 ft. from the right
trailer wall to the right quarter, center, and left quarter, respectively.
2' Code: K--break; B--bruise; S--scuff; N--no injury; C--crack.
3/ After settling.
Wall
Wall
Table 9.
Celery Petiole Damage
Bulk Handling Celery Test No. 2 -- Depth 4 ft.
/Location in Traile
Location in Trailer-
Left quarter
Center
Right quarter
2/
B K,B-
K,B 2K,B
K,B
B,S
N N
B,S K
K N
K,S N
K K
K,B K,2B
1/ Trailer was 7 ft. wide. Dimension location of samples were 1-3/4, 3-1/2 and 5-1/4 ft. from the right
trailer wall to the right quarter, center, and left quarter, respectively.
2/ Code: K--break; B--bruise; S--scuff; N--no injury; C--crack.
3/ After settling.
Depth
Level
Inches
393/
39-1
32
24
12
6
.7 '. .
K,B
N
Wall
K,B
B
K,S
S
N
K
K
N
K
B,S
2B
2B
S
S
N
N
N
K
K
K
2K
K,L,S
K,B
3K,B
2S,B
B
K,B N
S N
Left uarter Center--- --- --- --
Table 10.
Bulk Handling Celery Test
Percentage of Stalks with Marketable Petioles Cracked or Broken
1/
Level-Inches-
Test ITo.
Depth-Feet
2
3
4
6
1 2
2/
54- 29
54 50
46 42
12
1 2
71 33
58 38
46 50
24
1 2 :
62 33
62 38
54 38
36
1 2 :
46 21
42 42
48 Average
1 2 : 1 2
62 32
S 55 37
42 33 : 46 41
Before settling.
Each figure represents a sample of 24 stalks.
Pagel5
The occurrence of petiole breakage was high partially because a large number
of the petioles were broken during the cutting operation. Thi3 was particularly
true in Test No. 1 which was conducted in the morning when the relative humidity
was 100 percent and the celery was wet and extremely brittle. Although bruises
and scuffs were noted, their occurrences were not considered especially critical.
It was concluded from this study that celery can be bulk transported if
properly handled and that the depth to which it is stacked, up to four feet, does
not exert a critical effect on the amount of damage incurred during transporting.
In future studies, attention will be devoted to obtaining less damage than was
noted in this test.
MECHANICAL HARVESTER STUDIES
After initial information on the physical properties of celery was obtained
in 1964, work was begun on the development of the functional specifications for a
mechanical harvester.
A successful mechanical harvester must be capable of performing three basic
functions: (1) cut the plant from its root system, (2) remove the excess top from
the stalk, and (3) elevate the stalls to a height at which it can be further
manipulated.
The cutting consists of severing the aerial portion of the celery plant from
the root in the proximity of the soil surface. It is complicated by (1) the
trade's preference for packaged celery which has one-eighth to one-fourth inch of
root remaining on the stalk, and (2) the variability of the location of the base
of the petioles because of the inherent characteristics of a transplanted crop.
This situation makes it difficult to preset a cutting mechanism at some pre-
determined height and expect to maintain a uniform length of cut for the full
length of a field. The general physical appearance of the base of the stalk after
it has been harvested must also be considered. Since most of the celery produced
in Florida is grown on organic soils, moving or rotating devices used to cut the
celery often pick up soil particles and imbed them in the stalk. This leaves a
black stain which cannot normally be removed by washing.
1964 Studies
Several root stem cutting mechanisms were considered;
respective advantages and disadvantages, are listed below:
Cutting Mechanism
1. Reciprocating blade
2. Circular saw
3. Band Saw
4. Disk Coulter
5. Stationary knife
blade
Advantages
a. Good cutting action
a. Good cutting action
b. Easy to drive
c. High life expectancy
a. Excellent cutting action
a. Good cutting action
b. No drive mechanism
required
c. Long life expectancy
a. Fair cutting action
b. No drive mechanism
required
c. Long life
these along with their
Disadvantages
a. Difficult to drive
b. Accumulation of debris
and plant residues
a. Accumulation of plant
residues on rotating
parts
a. Difficult to position
at soil level
b. Short blade life
a. Some soils may dull
cutting edge
a. Frequent resharpenings
Page 16
Because of the apparent advantages of the disk coulter, it was selected as
the first cutting mechanism. In order to utilize the self-drive feature of the
coulters, the root cutting operation was divided into two parts: (1) a gross
initial cut and (2) a smooth final cut one-quarter inch below the base of the
petioles. The 1964 harvester was equipped with a set of disk coulters with the
upper edges tilted outward at 45 degree angles, which ran paralled to the line of
travel and converged two inches below the soil level.
These coulters left a wedge-shaped root on each plant which maintained the
alignment of the plant as it slid between a set of guides. The wedged portion of
the root was to be retrimmed later as the outer petioles came in contact with a
reference plane.
Initial plans included the use of a band saw, mounted at the rear of the
machine, which would retrim the root and cut the top to a suitable length for
packaging. Although the disk coulter system worked satisfactorily, the difficulties
encountered with the use of the band saw and other root retrimming devices on the
field machine forced abandonment of the two-part root cutting system.
The elevating system for the 1964 model consisted of a set of spring loaded
B-section V-belts which gripped the celery approximately 10 inches above the soil
level. These belts were inclined so that the celery was elevated approximately
two inches per foot as it moved through the harvester.
Stalk damage was assessed at speeds of 0.4, 0.6, and 0.8 miles per hour,
which was equivalent to 55, 82, and 110 plants per minute. The results of this
study are summarized in Table 3. Although the number of stalks damaged ranged
from 48 to 64 percent, the severity of the damage was slight and most of this
occurred on petioles which would normally be removed before packing. However,
most of the damage was caused by the grip belts and it was felt that efforts
should be made in subsequent tests to reduce this damage.
Table 3. Affect of Harvester Speed on Celery Damage Harvester Equipped with
V-Belts
Speed o stalks with damaged petioles
1. 0.4 mph. 64
2. 0.6 mph. 58
3. 0.8 mph. 48
1965 Studies
As a result of the 1964 studies the mechanical harvester was modified to in-
clude the use of a stationary knife blade for the root cutting operation. It was
established that the knife blade could be used in conjuction with a quick response,
manual depth control to cut the root stem at the proper location. Subsequent tests
indicated that less than one percent of the celery cut in this manner required re-
trimming. According to the results of studies on the force required to cut the
root stem, a blade placed at an angle of 60 degrees with a normal to the line of
travel required the smallest force to cut while fulfilling the geometric require-
ments of the harvester. A blade one-eighth inch by two and one-half inches located
near the front of the harvester produced a smooth, clean cut without displacing
the root with respect to the soil (Figures 4 and 5).
The top trimmer consisted of a rotating one-quarter inch thick steel disk 15
inches diameter, with six serrated edge mower blades attached to the periphery.
This trimmer rotated at 2,000 revolutions per minute and was located 15 inches
above the root cutting knife (Figures 4, 5, and 6). This space relationship
produced a trimmed plant of correct length for packaging.
Page 17
Since the grip belts used in 1964 caused undesirable dnmage and failed to
control the celery positively, the 1965 harvester was equipped with a set of flat
belts with control cleats attached (Figures 3, 4, 6, and 7).
Three types of control cleats were considered in the laboratory before a
decision was made with regard to the choice for the harvester. The cleats consider-
ed along with a physical description of each are as follows:
1. B-section V-belt riveted to the belt at two and one-half inch spacings,
2. Hollow neoprene tubing, with inside diameter of 0.62 inches and outside
diameter of 0.75 inches, also riveted to the belt at two and one-half
inch spacings,
3. Solid neoprene strips of rectangular cross section with a thickness of
0.3 inches and width of 1.5 inches, riveted to the belt at two and
one-half inch spacing.
Each of these cleats was tested by running a sample of 25 plants between the grip
belts with the appropriate cleat in place. Damage to the celery stalks was deter-
mined immediately after treatment and again after seven days storage at 360F. The
results of this study indicated that there was no significant difference between
treatments and that there was an insignificant amount of damage caused by any of the
treatments. It was concluded that any one of the three types of cleats could be
used. The V-belt section was selected because of its durability.
The grip belts were designed so that only the idler on the slack side of each
belt would be spring loaded. Idlers on the tension side of the belts were spaced
alternately, eliminating the possibility of crushing damage which might occur
between two opposing idlers. See Figure 5.
As can be seen in Figure 5, the celery plant is firmly gripped by the belts
before the plant is engaged by the root stem cutter or top trimmer. The belts
travel at ground speed so that the plant has no horizontal movement during the
cutting operation. This relationship has proved to be very successful, especially
under wet field conditions where the force to cut the root often exceeds the force-
required to remove the plant from the soil. Additional types of belts will be
tested in 1966.
The harvester was run at two forward speeds: 45 feet per minute and 64 feet
per minute. These speeds were equivalent to 92 and 134 plants per minute respecti-
vely. These harvesting rates were equivalent to that of the hand harvesting rates
observed in one of the principal celery growing areas during the 1964 season. At
that time three hand harvest operations were observed. The average number of hand
cutters was 20 per mule train. The average rate of hand harvest observed was 106
plants per minute per machine or 5.3 plants per minute per man. Randolph (3) re-
ported that under some circumstances a hand cutter crew can produce 300 crates per
hour which is equivalent to approximately 150 stalks per minute. Therefore it can
be concluded that a single-row mechanical harvester operating at 64 feet per
minute can closely approximate the cutting rate of 16-20 hand cutters.
An analysis was made to determine the damage characteristics of both the hand
and mechanical harvesting systems. A random sample was obtained from the horizontal
wing conveyors on each of two mule trains. Celery at this point has been hand cut
and has had the unmarketable petioles removed. This is a somewhat biased sample,
since the hand cutter normally strips damaged petioles from the plant before placing
it on the conveyors. However, it is at this point that celery from the mechanical
harvester can best be compared with that of the present system. Although 34 percent
of the stalks exhibited some amount of damage from hand cutting, the damage ranged
from a slight scuff to broken petioles, there were only four stalks in the hand cut
sample which would be classified as unmarketable. However, of the plants harvested
mechanically none was classified as unmarketable and the amount of unsightly damage
was less than 17 plants per hundred. Page 18
An elevating system was constructed for conveying the plants from the ex-
perimental harvester to a transport vehicle. Two opposing, soft backed, V-Belts
grip the stalks at the top as they are released from the cleated grip belts but
are still in an upright position. The elevating system is connected to the rear
of the mechanical harvester and has a 30 degree incline.
The pressure exerted on the tops of the trimmed stalks by these V-belts often
caused unacceptable damage. The elevating system was changed so that the stalks
can be gripped above the trimming height. This method of gripping stalks prohibits
top trimming as presently performed on the experimental harvester. Studies are
presently being conducted to determine the best method for gripping and topping the
stalks. Research will also be devoted to developing a method for handling the
celery as it is released by the elevating belts.
SUMMARY
A study has been conducted to determine values for most of the physical pro-
perties of celery required for the development of a mechanical harvesting-handling
system. These properties include plant profile, weight, h!.igLt, root stem diarleter,
center of gravity, coefficient of static and sliding friction, resistance to damage,
and the force required to cut the root stem.
An experimental, one-row, mechanical harvester has been developed which will
harvest at the rate of 134 plants per minute. It will cut the root stem and top
the celery plant at the desired length for pac sL';i-~ The re Isu it of d6.mage
studies indicated that the mechanically harvested plants exhibit l.css damage than
hand harvested plants. An elevating-bulk handling system for conveying and trans-
porting the mechanically harvested celery plants is being devo.loped.
Two packing systems for utilizing a mechanical harvester have been proposed.
Satisfactory progress has been made in the eo-'elopment of mechanical celery
harvesting to merit future studies in bulk handling, machine stripping, machine
sizing, and automatic packaging in addition to refinements on the experimental
harvester.
REFERENCES
1. Beeman, J. F., W. W. Deen, Jr., and L. H. Halsey. Development of a mechanical
celery harvester. ASAE Paper No. 65-622. Presented at 1965 Winter Meeting,
Chicago, Ill. Dec. 7-10.
2. Florida Agricultural Statistics, Vegetable Summary 1964. Fla. Dept. of
Agriculture, Tallahassee, Fla.
3. Randolph, John W. Mobile celery harvesting-packing combine. Agricultural
Engineering 37: 755-758, 1956.
4. Spurlock, A. H. Unpublished Progress Report, 1965. Dept. of Agricultural
Economics, Univ. of Fla., Gainesvjlle, Fla.
EES 66-8
350 Copies
Gage Wheels
FIG. 4 Schematic of a single-row mechanical celery harvester. (Side view)
FIG. 5. Schematic of a single-row mechanical celery harvester. (Top view)
11/1114160
x/.
FIG. 6. Tractor mounted celery harvester operating under
field conditions.
1
.
C '''F .i
~r r
r
FIG. 7. Rear view of celery harvester showing
and hydraulic controls.
grip belts
|