- Email: [email protected]
Development of prototype harvester for head lettuce
Engineering in Agriculture, Environment and Food xxx (2014) 1e8
Contents lists available at ScienceDirect
Engineering in Agriculture, Environment and Food journal homepage: http://www.sciencedirect.com/eaef
Development of prototype harvester for head lettuce Van Nguyen Nang*, Suguru Yamane Shizuoka Prefectural Research Institute of Agriculture and Forestry, 678-1 Tomioka, Iwata City, Shizuoka Prefecture 438-0803, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 July 2014 Received in revised form 1 September 2014 Accepted 11 September 2014 Available online xxx
Method of manual removal of produce from the field has been a bottle-neck for lettuce production in Japan. In order to reduce the production cost of this fresh commodity, a prototype harvester has been developed for head lettuce production. The harvester consists of a cutting component to slice the lettuce head at the desired location and a lifting component that transports the harvested produce from the cutting site onto elevating conveyor and trimming station. A cutting component with reciprocating blade was proposed. Laboratory tests were performed to verify ability of reciprocating blade to slice lettuce stump at forward cutting speed of 0.1 m/s, reciprocating stroke of 18 mm, and different reciprocating frequencies of 2, 4 and 6 Hz. In addition, power requirement for reciprocating the cutting knife as slicing two lettuce stumps was measure. Tests in lettuce fields were also conducted at different working speed to investigation the cutting and lifting performances of the harvester mounted with the reciprocating-blade cutting component. The results of laboratory tests indicated that the cutting component could smoothly cut lettuce stumps and the maximum cutting torque and cutting power requirement were 0.73 Nm and 27.7 W, respectively at 6 Hz reciprocating frequency. Field test results showed that the harvester could cut and lift the lettuce heads without damaging and blemishing the produce at working speed of 0.04 m/s and the commercial head percentage was 94.5%. At higher working speed of 0.08 m/s, the head damage rate was 12.8% reducing the percent of commercially accepted heads to 87.2%. © 2014, Asian Agricultural and Biological Engineering Association. Published by Elsevier B.V. All rights reserved.
Keywords: Lettuce Harvester Cutting component Lifting component Harvesting performance
1. Introduction Lettuce, (Lactuca sativa L.), is native to the Mediterranean area and was introduced to Persian cuisine around 4500 B.C. It arrived in Japan during Heian period (794-1185 A.C.), but did not become an important commodity until the Western cuisine changed the Japanese food culture in 1980's. About 20,900 ha of the lettuce are grown year round in Japan, and 30% of which is produced during winter time in some production areas such as Nagasaki, Kumamoto, Kagawa, Shizuoka, and Ibaraki prefectures. There are approximately 791 ha of lettuce grown in Shizuoka including head and leaf lettuces. Today, Shizuoka ranks third, following Kagawa and Kumamoto prefectures, in production of winter head lettuce. The fresh produce is the prefecture's leading cash crop averaging more than 5.2 billion yen in value. Winter head lettuce is grown in paddy fields with four rows per bed cultural system using overall mulching and plastic-tunnel. The lettuce seedlings are transplanted * Corresponding author. Tel.: þ81 538 36 1551. E-mail address: [email protected] (V.N. Nang).
in a zigzag pattern with hill distance and inter row space varying from 27 to 30 cm. While main farming practices on lettuce production such as soil preparation, transplantation, and insect pest management have been mechanized, lettuce is still harvested by hand-cut in a stopped and squatting posture using extensive labor. Furthermore, the operation is often conducted in two times due to the nonuniformity of lettuce growth within plastic film tunnels. The required working hours to produce 1 ha of lettuce is 670 h, 32% of which involves in harvesting work. Although manual labor is still preferred for freshmarket vegetable harvesting operations because of the delicate nature of the products, the need of mechanization depends upon the availability of human labor and the level of industrialization within each country. Although Japan is one the most advanced countries, its population is decreasing and aging and only 2% of the Japanese population is engaged in production agriculture now. This resulted in a severe lack of human labor for farming operations. In these circumstances, harvest work has been a bottle-neck of cost reduction and acreage expanding for lettuce production in Shizuoka prefecture.
http://dx.doi.org/10.1016/j.eaef.2014.09.002 1881-8366/© 2014, Asian Agricultural and Biological Engineering Association. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
2
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
The incentive for mechanization of lettuce harvest depends on the available of harvesters that can achieve economies through substitution of capital for labor. Since Japanese lettuce industry is small in terms of numbers of machines and equipment that can be used, agricultural machinery companies are reluctant to invest in research and development. Therefore, research projects have been initialized by central or local governments to produce prototype lettuce harvesters which commercial firms use to adapt to commercial uses. The development concept was to replace the hand selector-cutter-trimmers with the machine. Two prototypes of once-over harvester were developed and tested for summerclipping head lettuce grown one row per bed (Suzuki et al., 2003; IAM-BRAIN, 2003). The first prototype was developed at Agricultural Experiment Station of Nagano prefecture. The machine cut the lettuce by a rotating serrated disc and lifted it to leaf eliminating elevating conveyor at which hand trimmers removed the wrapper leaves. The trimmed lettuce then was delivered to the accumulating belt to dissipate the latex using water washing nozzles and packed in the cartons. The later prototype that was developed by IAMBRAIN consisted of two horizontal orientating discs, followed by an oscillating circular-sector cutting blade and a tilted floating conveyor. A cutoff unit was mounted at the rear end of the machine. However, none of these experimental harvesters were commercially accepted because of high labor input requirement. The main bottleneck of these harvesters was hand trimming and latex dissipation. Consequently, an experimental once-over lettuce harvester has been constructed and tested for four-row furrow-irrigated cultural system at Shizuoka Prefectural Research Institute of Agriculture and Forestry. The objectives of the harvester development are to mechanize the lettuce harvesting operation which will helps prefectural lettuce growers to reduce production cost as well as to expand lettuce acreage by using the abandoned arable. The harvesting concept is to cut the lettuce head and lift it onto packing belt. The hand trimming and latex dissipating will be conducted at the storage facility. The experimental harvester consisted of a chassis with rubber crawler running gears, a cutting component, a lifting component, and a packing station. The output product of this harvester is undamaged and unblemished heads with straight stump cut and 2 to 6 wrapper leaves, and the rate commercialized heads over harvested heads is more than 95%. In this paper, the development of a cutting component with a reciprocating blade and a lifting component is mainly described and results of laboratory and field tests are also reported to evaluate their performance. 2. Materials and methods 2.1. Conventional practice of lettuce harvest Japanese winter lettuce transplantation starts at the end of September and continues to the beginning of February. Harvesting
takes place primarily from November through April. When temperatures cool down, lettuce rows are covered using plastic film tunnels to minimize the depressed effect of cold weather on the plant growth. However, this farming practice has involved much in the growth difference between the inner and outer rows of lettuce due to the nonhomogeneity of the temperature distribution inside the film tunnels. Once-over harvest of lettuce head is applied before January (Fig. 1a) and after that selective harvest is conducted by cutting two inner rows first (Fig. 1b and c). It takes about a week to do the second harvest of the outer rows. After the lettuces are selected and cut using a sharp cutting implement such as a knife, they are trimmed with a certain number of wrapper leaves left. Lettuce in Shizuoka is shipped using two packing techniques: naked pack or film wrap. Field packaged and shipped lettuce is cut at about the middle leaf, and trimmed to leave three wrapper leaves, then packed in the carton or plastic containers. Each container contains two layers of 12e18 heads depending on the head size. The top layer packed stump-up is washed to dissipate latex sap which excluded from the stump cut before shipping. For film-wrapped lettuce, heads are cut under the last basal leaf, trimmed to remain three to five wrapper leaves, then packed in containers and transported to storage facilities. The produces are then trimmed again to remain one wrapper leaf, dissipated the latex sap and wrapped with cellophane at the facilities before shipping. After harvest, the beds with plastic mulch will be used for further planting of sweet corn or even for the second transplantation of lettuce. 2.2. Design concept of the experimental harvester Generally, manually harvested lettuce is a mature, undamaged and unblemished head having the stump cut straight across to remain a desire number of wrapper leaves. Because winter head lettuce in Japan is grown inside plastic film tunnels, selective mechanical lettuce harvest is technically unfeasible. Some practices such as utilizing starter dressing or green house culture have been studied at the Shizuoka Prefectural Research Institute of Agriculture and Forestry to improve the growth uniformity of winter lettuce that can minimize the harvest lost due to immature heads if once-over harvest is applied. Consequently, the experimental lettuce harvester must meet the following requirements: (a) to mechanize once-over harvest of multi-row winter lettuce with efficiency four times higher than hand cut; (b) to injure and blemish the heads as less as possible; (c) not to damage the mulch; (d) to improve the working posture of workers; (e) for naked pack technique, lettuce heads will be trimmed to remove unwanted wrapper leaves, packed and dissipated the latex sap at the packing station;
Fig. 1. Conventional harvesting practices of head lettuce in Shizuoka prefecture.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
(f) for plastic film wrap technique, lettuce heads will be trimmed to remove unwanted wrapper leaves and packed. (g) harvested lettuce will be delivered out of field using a vegetable carrier. 2.3. Development of cutting component 2.3.1. Lettuce cutting methods Cutting unit is the most important component of a lettuce harvester and many previous studies have paid attention on design and development of a high accuracy cutting method for head lettuce. Suzuki et al. (2003) used a stationary knife to cut lettuce plant grown in one row per bed system. The knife was positioned near the rear end of front end tilted and angled gage discs which was connected to a parallel link. This cutting unit could cut lettuce stump at maximum harvester speed of 0.27 m/s without injuring lettuce heads. Researchers at The Institute of Agricultural Machinery (IAM-BRAIN, 2003) also developed an experimental lettuce harvester having similar gaging system for a powered arc-shaped cutting knife. The cutting unit could cut one row per bed lettuce at harvester speed of 0.06 m/s without cut-into heads and yield acceptable range of wrapper leaves. However, the cutting accuracy decreased as working speed increased. The rate of cut-into heads was approximately 20% at speed of 0.16 m/s. There have been several methods for cutting lettuce head in a two rows per bed system. Shepardson et al. (1974) developed a harvester for lettuce which was seeded two rows per bed. The author investigated the field performance of some cutting devices using stationary knife, band saw, blade saw, overlapping discs, and pinch-off discs. The front end height of the cutter heads was controlled by powered orientating discs. However, no devices were found accurately cut the stump of lettuce to ideally leave four outer leaves. Heads needed to be recut to obtain the quality of cut. Furthermore, dirt problem also occurred because low-positioned orientating discs scooped up some of the muck soil and the band saw or pinch-off discs scrubbed some soil into the stump cut. An investigation was made to determine if the 2-row lettuce grown on a specially flat bed could be cut accurately by a stationary knife (Adrian et al., 1976). The knife was adjustably mounted on a gage wheel that ran on the center of the bed. The height variation
3
Table 1 Specifications of cutting blade. Item
Specification
Material Hardness Dimensions (width thickness) Edge shape Sharpening angle Weight with sliding frame
Stainless steel (SUS-MS) HRC58 ± 2 1380 60 3.2 mm Smooth, single bevel 11 4260 g
between the row lines and the center of the bed was 2.5 cm. Result of cutting experiment indicated that cutting accuracy was greatest expectable with only 5% of cut-into heads. However, accuracy of this cutting method would depend on changes in cultural practices so substantial as to be economically unfeasible. Authors suggested that a reference rather than the bed center should be used. Water jet was also considered as a lettuce cutting method with advantages over other mechanical cutting methods such as minimization of acceleration and inertia problems and optimization of space requirement for cutting mechanism (Schield et al., 1973). Field test result showed that lettuce stumps could be cut with a water jet but no data on the cutting accuracy was reported. Furthermore, problem of controlling the nozzle path occurred as the harvesting module advanced down the row. The method also had potential problems such as unclean cut, dirt, high power requirement, water supply and dangerous high-pressure nozzle. 2.3.2. Design of cutting component Based on the available cutting methods together with the lettuce cultural system in Shizuoka, we have been investigating the cutting accuracy of a cutting component (Fig. 2) that uses a reciprocating blade to slice lettuce stems at a predetermined cutting height of the blade with respect to soil bed surface. This cutting method was chosen because it can simultaneously cut the four rows of lettuce grown on a fairly flat bed system, which is used in lettuce production areas in Shizuoka (Nang et al., 2013). In addition, the coherent problems of mechanical cutting mechanism, which are high acceleration and inertia, heavy weight, large space requirement for positioning blades, actuating devices and linkages,
Fig. 2. Schematic view of reciprocating-blade cutting component.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
4
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
can be eliminated by optimal design for weight and dimension of the cutting component. The structure of reciprocating-blade cutting component was described in details in a previous report (Nang et al., 2013). In this development, the cutting bade that had a smooth single beveled edge was made of 3.2 mm thick stainless steel plate and sharpened at an angle of 11. Its' detailed specifications are reported in Table 1. The effects of blade thickness, edge shape (scalloped or serrated) and sharpening angle on the cutting quality and torque requirement will be taken into account in a future study. The cutting component is positioned at a front portion of the harvester and adjustably mounted on a frame of the lifting component. The cutting and lifting component rides on powered gage wheel having five rollers that are positioned behind the cutting blade and between the lettuce rows to follow the bed contour. The gap between rollers is large enough so that the gage wheel does not run over the cut stump left on the bed surface. The rear end of lifting component frame is mounted on harvester main frame by a universal joint that let the gage wheel follows the bed contour even the left and right furrows have different depth. 2.3.3. Lettuce stem cutting experiments with reciprocating bade Cutting experiments were conducted to verify performance of reciprocating-blade cutting component at different combinations of reciprocating stroke, frequencies and harvesting speed. In the previous study (Nang et al., 2013), the measurements of lateral cutting resistance, required cutting torque and power for slicing two lettuce stems using a stationary blade were made at reciprocating frequency and stroke of 4 Hz and 18 mm, respectively and at different forward speeds. In this report, power requirement for cutting two and four lettuce stumps was also measured as a reference for choosing driving motor of cutting blade at different reciprocating frequencies. Experimental apparatus is depicted in Fig. 3. A movable carrier that simulated the soil bed of lettuce carried two or four stumps (average stem diameter of 40 ± 5 mm) of over-mature lettuce plants through the stationary cutting component at forward speed of 0.1 m/s. The cutting blade was reciprocated at a stroke of 18 mm, and rotational speeds of 120, 240, and 360 rpm or blade reciprocating frequencies of 2, 4, and 6 Hz, respectively. Torque for driving cutting blade together with rotational speed of crank shaft was measured using a torque meter (TQR-300N,
Nikkei), an encoder (E6C2-CWZ6C, Omron). Crank shaft angle was detected by a proximity sensor (TL-W5E1 5M, Omron). The measured data were recorded through a 10 Hz low-pass filter at a sampling rate of 1 KHz using a data acquisition system (Memory Hicorder 8842, Hioki). Each experiment was replicated 5 times. 2.4. Development of lifting component 2.4.1. Methods for lifting cut lettuce heads from ground Lifting component takes cut produce from the back of cutting component and carries them onto the horizontal packing conveyor without injuring the heads or remained wrapper leaves. Shepardson et al. (1974) used rod-type chain belting to elevate cut lettuce and eliminate dirt at an inclined angle below 28 . This component could not convey heads in any given orientation and it delivered debris along with lettuce. The authors also used two side gripping resilient belts to receive cut lettuce and maintain its orientation for transfer of the produce to the elevating conveyor consisting of top and bottom belts. Excess cut off leaves, undersized heads and debris dropped free from this conveyor. Rubber-fingered belts (Lenker et al., 1973), fan-shaped sponge conveyors (Suzuki et al., 2003), and wave-formed belts (IAM-BRAIN, 2003) were used to grasp lettuce by their sides just before they are cut and then lift the cut heads to the trimming conveyor. The introduced methods for lifting cut produce onto trimming platform were applied for one row per bed system or two rows per bed system with large enough row spacing for positioning lifting belts. 2.4.2. Design of lifting component Because Shizuoka winter lettuce is grown four rows per bed with condensed zigzag pattern, the principle of lifting cut heads by their left and right side using two-side gripping belts appeared unfeasible. Therefore a top and bottom belts were used to elevate lettuce to the packing conveyor (Fig. 4). The bottom belt that travels upward is 1100 cm wide and is made of urethane and polyester. The top belt is a rod-type chain belting system, which consist of four wave-formed urethane belts fixed to the rods at a pitch of 20 cm. The wave-formed part is 150 cm wide, 20 cm high and positioned in zigzag pitch pattern. The arrangement of wave-formed parts significantly affects the performance of lifting component so their optimal size will be considered in the next step of machine development. The top belt travels downward and resiliently
Fig. 3. Experimental apparatus for cutting lettuce stems.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
5
Fig. 4. Schematic view lifting component mounted on the experimental lettuce harvester.
compresses the heads as they are cut then together with the bottom belt gasp and lift the produce to the packing station. The contact area between lettuce head and the wave-formed belts are covered with 5 mm-thick sponge sheets to reduce impact force of the belts on the heads as well as to increase friction between them hence improve the lifting performance of the top belt. The lifting component conveys cut heads at an inclined angle that varies from 28 to 33 depending on bed height. During harvesting operation, the front end of lifting component rides on the five-roller gage wheel. In travelling condition, the cutting and lifting components are raised up by two arms that are actuated by an electric powered hydraulic cylinder. The experimental harvester is equipped with 2 electric generators with total power of 2.5 kW and with rubber tracks, which allow a regular forward movement on any kind of terrains. The track-center distance can be varied from 1.6 to 2 m. In the final step of development, the harvester will be equipped with an automatic pilot apparatus so that only a packing crew is on board during harvesting operation. 2.5. Field performance of developed harvester Field tests were conducted to evaluate the performances of reciprocating-blade cutting component and lifting component. Winter lettuce was grown on fairly flat beds having width of 2 m and height of 20 cm in an experimental field of Shizuoka Prefectural Research Institute of Agriculture and Forestry (Fig. 5). Growth conditions of lettuce before harvesting are reported in Table 2. The
tests were performed at harvesting speeds up to 0.24 m/s with cutting stroke of 18 mm and blade reciprocating of 4 Hz. Although the optimal cutting height to remain 2 to 3 wrapper leaves on the cut head was approximately 16 mm, initial height of cutting blade was set 10 mm above bed level for every test in order to minimize the number of heads with too few wrapper leaves. Two packing workers walked behind the harvester on the bed surface to pack cut heads into plastic containers. Working functions of cutting and lifting components were investigated for every test while quality of lettuce in terms of remained wrapper leaves, damaged rate, dirt problem were measured to evaluate the accuracy of cutting component as well as the delivery performance of the lifting component. 3. Results and discussions 3.1. Results of cutting experiments The results of cutting experiments indicated that the reciprocating cutting component could smoothly cut lettuce stumps. Sample data of measured torques for cutting lettuce stumps at different blade reciprocating frequencies are displayed in Fig. 6. The measured total torque was a sum of unloading torque for reciprocating the cutting blade in noncutting condition and torque for cutting lettuce stem. The unloading torque was synchronized with the total cutting torque with respect to the crank angle. A sample of synchronized data of one rotation of the crank shaft in which the maximum total torque occurred is depicted in Fig. 7. The unloading torque had a wave form as a function of crank angle with fluctuation amplitude. It changed direction 3 times in each rotation of the crank shaft and fluctuation pattern was similar for both the forward and return strokes. The torque increased in the
Table 2 Physical characteristics of harvested lettuce.
Fig. 5. Overview of developed experimental harvester.
Variety Total plant weight, g Total plant height, cm Number of wrapper leaves Head weight, g Head height, cm Head transverse diameter, cm Head conjugate diameter, cm Optimal cutting height, mm
M-WRAP-231 599 ± 128 18 ± 1 12 ± 2 396 ± 64 16 ± 1 16 ± 2 14 ± 2 16 ± 5
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
Measured total torque (N.m)
6
1.2 Reciprocating frequency 1
2Hz
4Hz
6Hz
0.8 0.6 0.4 0.2
0 -0.2 0
0.5
1
1.5
2
2.5
3
Time (s) Fig. 6. A sample of measured toque for cutting four lettuce stumps.
first half of both forward and return strokes and then decreased at the second halves. It reached maximum and minimum values at the return stroke. This was contributed to the acceleration and inertia of the cutting blade. The mean value of maximum unloading torques increased with increasing rotational speed of crank shaft and was 0.16, 0.35 and 0.49 Nm for reciprocating frequencies of 2, 4 and 6 Hz, respectively (Fig. 8). The torque measured during lettuce stump cutting was greater than that in unloading condition and reached a peak several times before the lettuce stems were totally cut. Theoretically, the amplitude of the total torque depends on the reciprocating mass, the cutting resistance and the moment that the cutting blade started slicing the lettuce stems. However, the maximum value of the total torque always occurred at the end of the forward stroke. The mean value of maximum total torques was proportional to rotational speed of crank shaft. In the case of 2-stump cutting, it was 0.35, 0.56 and 0.74 Nm corresponding to reciprocating frequencies of 2, 4 and 6 Hz, respectively, while that was 0.70, 0.94 and 1.06 Nm for cutting 4 stumps, almost twice greater than cutting 2 stumps (Fig. 8). Meanwhile, the torques for cutting 2 or 4 stumps were slightly increased with increasing reciprocating frequency. The mean of maximum values of cutting torques was 0.28, 0.31, 0.32 Nm in the case of 2-stump cutting and 0.60, 0.67, 0.72 Nm in the case of 4stump cutting with respect to reciprocating frequencies of 2, 4 and 6 Hz, respectively. The required power for cutting lettuce stumps was calculated from total torque and rotational speed of crank shaft. It was 4.3, 14.0, 27.7 W for cutting 2 lettuce stumps and 8.9, 23.6, 39.8 W for cutting 4 lettuce stumps corresponding to reciprocating frequencies of 2, 4 and 6 Hz, respectively (Fig. 8).
3.2. Results of field tests Field test results indicated that the experimental harvester could harvest lettuce at ground speed up to 0.24 m/s. Satisfactory operation with packing crew of 2 workers only were obtained at harvesting speeds under 0.10 m/s. For higher speed, more packing workers or bulk packing of harvested lettuce would be necessary. The cutting component properly functioned at reciprocating stroke of 18 mm and frequency of 4 Hz and successfully cut lettuce plants (Fig. 9). At these reciprocating parameter, vibration of the component was small so that it is unnecessary to reduce the acceleration and inertia of the cutting mechanism by using counterweight for crank shaft. Generally, cutting blade reciprocates just above the mulch so dirt problem was not observed and clean cuts of lettuce stumps were also obtained. The gage wheel could follow the bed contour but it was not capable of controlling the height of cutting blade on irregular bed surface because the distance from front end of the blade to the axle center of the gage wheel is rather long, approximately 30 cm. The diameter of the gage wheel should be reduced to shorten this distance in order to improve the ability of cutting height control. The cutting blade should be positioned near the front end of the bottom belt of the lifting component to facilitate the cut head delivery. The lifting component could also successfully elevate the cut heads to the packing belt (Fig. 10). The font end of bottom belt should be position near the bed surface to smooth the head delivery. Some bruises or breakage of wrapper leaves were occurred as the heads were elevated from the cutting blade to the bottom belt. However the heads were salvaged by further trimming of unwanted wrapper leaves. The traveling speed of the top and
Fig. 7. A sample of synchronized toque for cutting four lettuce stumps.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
Total torque Cutting torque Power requirement
1.2
90
0.8
60
0.4
30
0.0
0 0 2 4 6 8 Reciprocating frequency (Hz)
120
1.6 Cutting 4 stumps
Torque (Nm)
Torque (Nm)
Cutting 2 stumps
1.2
90
0.8
60
0.4
30
0.0
0 0 2 4 6 8 Reciprocating frequency (Hz)
Power requirement (W)
120
Power requirement (W)
1.6
Unloading torque Total torque Cutting torque Power requirement
7
Fig. 8. Measured and computed torque for different reciprocating frequency.
Fig. 9. Lettuce stumps left on the bed (left) and harvested heads (right).
bottom belts were synchronized with harvesting speed to minimize the head and wrapper leaf injures due to relative motion between lettuce and lifting component. The stumps set on the bottom belt and the head was held by the resilient top belt. The wrapper leaves. The component conveyed mature heads along with cut off leaves, undersized heads, undeveloped heads, and debris. Only commercialized heads were sorted on the packing conveyor and packed into plastic containers, while other was dropped freely on to the bed surface. Investigations of machine harvested lettuce quality were conducted for harvesting speeds of 0.04 and 0.08 m/s. The results are reported in Table 3. Because the pre-set of blade cutting height was lower than optimal cutting height, the percentage of acceptable
heads with 2e6 wrapper leaves was only 46.8 for harvesting speed of 0.04 m/s. However that increased to 73.1% at higher speed of 0.08 m/s. This indicated that the pre-set cutting height could not assure a desirable cutting accuracy with acceptable number of wrapper leaves. An active method of controlling the cutting height of blade during harvesting operation is required. The percentage of heads with less than 2 wrapper leaves, including the cut-into heads, was 2.9 and 10.6% for harvesting speeds of 0.04 and 0.08 m/s, respectively. This was partly due to variations between lettuce row line height, and partly due to the variations in length of lettuce stumps which resulted from nonuniformity of plant growth. There were some harvested heads with ribs of wrapper leaves that were bruised and/or broken at the gap between the cutting blade and the front end of the bottom conveyor or somewhere along the lifting component frame. However these heads were salvaged by the trimming cut. The head injure was due to the interaction between the top waveformed belts and the delivered heads that occurred when there was a difference in speed of the harvester and the lifting component. The rate of head injure increased with harvesting
Table 3 Quality of machine harvested lettuce. Commercialized Ground Number of Cut quality (%) head ratea speed investigated Too few Acceptable Too many Head heads (m/s) (%) leaves leaves injure 0.04 0.08
Fig. 10. Side view of lifting component in operation.
200 192
2.9 10.6
46.8 73.1
50.3 16.3
2.8 7.2
94.5 87.2
a Commercialized head rate: percentage of machine-harvested heads having greater or equal to two wrapper leaves without head injuries. Note that some injured cut heads would also have less than two wrapper leaves or vice versus.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
8
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
speed and was 2.8 and 7.2% for harvesting speeds of 0.04 and 0.08 m/s, respectively. The heads with number of wrapper leaves less than 2 or with head injuries were classified as unmarketable heads. Consequently, with a further trimming, the rate of commercialized head was 94.5 and 87.2% corresponding harvesting speeds of 0.04 and 0.08 m/s. Improvements for more accurate performance of cutting and lifting components are needed to obtain the targeted commercialized of harvested heads of over 95%.
2. Cut lettuce can be successfully elevated to the packing belt by the developed lifting component consisting of a bottom and top belts. 3. Maximum torque and power requirement for cutting 2 lettuce plants were 0.73 and 27.7 Nm, respectively. 4. For harvesting speed under 0.10 m/s, the experimental harvester could harvest lettuce with rate of commercialized heads ranging from 87.2 to 94.5%. 5. Accuracy of lettuce cutting of the experimental harvester was affected by the uniformity of soil bed profile and lettuce growth.
4. Conclusions
References
This study reports the development of a prototype harvester for winter-clipping head lettuce in Japan. The research and development of the individual components assembled in to an experimental lettuce harvester indicated the feasibility of mechanical harvesting of head lettuce. The prototype machine that was implemented with a reciprocating-blade cutting component could harvest lettuce with almost acceptable marketable rate at forward speed less than 0.1 m/s. Some conclusions could be drawn from the experimental work as following. 1. Lettuce stumps can be successfully cut by the developed cutting component with a reciprocating blade.
Adrian PA, Lenker DH, Nascimento D. Concepts and development of an experimental mechanical trimmer for crisphead lettuce. Trans ASAE 1976;19(1): 17e20. Institute of Agricultural Machine, Bio-oriented Technology Research Advancement Institution (IAM-BRAIN). Development of a Crisphead lettuce harvester. Annual Technical Research Report. 2003. p. 136e7. Nang V.N., Suguru Y., Cutting component development for a lettuce harvester, Annual International Meeting of American Society of Agricultural and Biological Engineers, Kansas City, Missouri, July 21eJuly 24, 2013, Paper No. 1316326132013. Schield M, Harriott BL. Cutting lettuce stems with a water jet. Trans ASAE 1973;16(2):440e2. Shepardson ES, Pollock JG, Rehkugler GE. Research and development of a lettuce harvester. Trans ASAE 1974;17(2):212e6. Suzuki H, K.Saito K, Shiokawa H, Sato K, Okada K, Mizutani K, et al. Lettuce harvester development (Part 4) e development of trial machine for practical use and field examinations. J Jpn Soc Agric Mach 2003;65(6):141e8.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
Contents lists available at ScienceDirect
Engineering in Agriculture, Environment and Food journal homepage: http://www.sciencedirect.com/eaef
Development of prototype harvester for head lettuce Van Nguyen Nang*, Suguru Yamane Shizuoka Prefectural Research Institute of Agriculture and Forestry, 678-1 Tomioka, Iwata City, Shizuoka Prefecture 438-0803, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 11 July 2014 Received in revised form 1 September 2014 Accepted 11 September 2014 Available online xxx
Method of manual removal of produce from the field has been a bottle-neck for lettuce production in Japan. In order to reduce the production cost of this fresh commodity, a prototype harvester has been developed for head lettuce production. The harvester consists of a cutting component to slice the lettuce head at the desired location and a lifting component that transports the harvested produce from the cutting site onto elevating conveyor and trimming station. A cutting component with reciprocating blade was proposed. Laboratory tests were performed to verify ability of reciprocating blade to slice lettuce stump at forward cutting speed of 0.1 m/s, reciprocating stroke of 18 mm, and different reciprocating frequencies of 2, 4 and 6 Hz. In addition, power requirement for reciprocating the cutting knife as slicing two lettuce stumps was measure. Tests in lettuce fields were also conducted at different working speed to investigation the cutting and lifting performances of the harvester mounted with the reciprocating-blade cutting component. The results of laboratory tests indicated that the cutting component could smoothly cut lettuce stumps and the maximum cutting torque and cutting power requirement were 0.73 Nm and 27.7 W, respectively at 6 Hz reciprocating frequency. Field test results showed that the harvester could cut and lift the lettuce heads without damaging and blemishing the produce at working speed of 0.04 m/s and the commercial head percentage was 94.5%. At higher working speed of 0.08 m/s, the head damage rate was 12.8% reducing the percent of commercially accepted heads to 87.2%. © 2014, Asian Agricultural and Biological Engineering Association. Published by Elsevier B.V. All rights reserved.
Keywords: Lettuce Harvester Cutting component Lifting component Harvesting performance
1. Introduction Lettuce, (Lactuca sativa L.), is native to the Mediterranean area and was introduced to Persian cuisine around 4500 B.C. It arrived in Japan during Heian period (794-1185 A.C.), but did not become an important commodity until the Western cuisine changed the Japanese food culture in 1980's. About 20,900 ha of the lettuce are grown year round in Japan, and 30% of which is produced during winter time in some production areas such as Nagasaki, Kumamoto, Kagawa, Shizuoka, and Ibaraki prefectures. There are approximately 791 ha of lettuce grown in Shizuoka including head and leaf lettuces. Today, Shizuoka ranks third, following Kagawa and Kumamoto prefectures, in production of winter head lettuce. The fresh produce is the prefecture's leading cash crop averaging more than 5.2 billion yen in value. Winter head lettuce is grown in paddy fields with four rows per bed cultural system using overall mulching and plastic-tunnel. The lettuce seedlings are transplanted * Corresponding author. Tel.: þ81 538 36 1551. E-mail address: [email protected] (V.N. Nang).
in a zigzag pattern with hill distance and inter row space varying from 27 to 30 cm. While main farming practices on lettuce production such as soil preparation, transplantation, and insect pest management have been mechanized, lettuce is still harvested by hand-cut in a stopped and squatting posture using extensive labor. Furthermore, the operation is often conducted in two times due to the nonuniformity of lettuce growth within plastic film tunnels. The required working hours to produce 1 ha of lettuce is 670 h, 32% of which involves in harvesting work. Although manual labor is still preferred for freshmarket vegetable harvesting operations because of the delicate nature of the products, the need of mechanization depends upon the availability of human labor and the level of industrialization within each country. Although Japan is one the most advanced countries, its population is decreasing and aging and only 2% of the Japanese population is engaged in production agriculture now. This resulted in a severe lack of human labor for farming operations. In these circumstances, harvest work has been a bottle-neck of cost reduction and acreage expanding for lettuce production in Shizuoka prefecture.
http://dx.doi.org/10.1016/j.eaef.2014.09.002 1881-8366/© 2014, Asian Agricultural and Biological Engineering Association. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
2
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
The incentive for mechanization of lettuce harvest depends on the available of harvesters that can achieve economies through substitution of capital for labor. Since Japanese lettuce industry is small in terms of numbers of machines and equipment that can be used, agricultural machinery companies are reluctant to invest in research and development. Therefore, research projects have been initialized by central or local governments to produce prototype lettuce harvesters which commercial firms use to adapt to commercial uses. The development concept was to replace the hand selector-cutter-trimmers with the machine. Two prototypes of once-over harvester were developed and tested for summerclipping head lettuce grown one row per bed (Suzuki et al., 2003; IAM-BRAIN, 2003). The first prototype was developed at Agricultural Experiment Station of Nagano prefecture. The machine cut the lettuce by a rotating serrated disc and lifted it to leaf eliminating elevating conveyor at which hand trimmers removed the wrapper leaves. The trimmed lettuce then was delivered to the accumulating belt to dissipate the latex using water washing nozzles and packed in the cartons. The later prototype that was developed by IAMBRAIN consisted of two horizontal orientating discs, followed by an oscillating circular-sector cutting blade and a tilted floating conveyor. A cutoff unit was mounted at the rear end of the machine. However, none of these experimental harvesters were commercially accepted because of high labor input requirement. The main bottleneck of these harvesters was hand trimming and latex dissipation. Consequently, an experimental once-over lettuce harvester has been constructed and tested for four-row furrow-irrigated cultural system at Shizuoka Prefectural Research Institute of Agriculture and Forestry. The objectives of the harvester development are to mechanize the lettuce harvesting operation which will helps prefectural lettuce growers to reduce production cost as well as to expand lettuce acreage by using the abandoned arable. The harvesting concept is to cut the lettuce head and lift it onto packing belt. The hand trimming and latex dissipating will be conducted at the storage facility. The experimental harvester consisted of a chassis with rubber crawler running gears, a cutting component, a lifting component, and a packing station. The output product of this harvester is undamaged and unblemished heads with straight stump cut and 2 to 6 wrapper leaves, and the rate commercialized heads over harvested heads is more than 95%. In this paper, the development of a cutting component with a reciprocating blade and a lifting component is mainly described and results of laboratory and field tests are also reported to evaluate their performance. 2. Materials and methods 2.1. Conventional practice of lettuce harvest Japanese winter lettuce transplantation starts at the end of September and continues to the beginning of February. Harvesting
takes place primarily from November through April. When temperatures cool down, lettuce rows are covered using plastic film tunnels to minimize the depressed effect of cold weather on the plant growth. However, this farming practice has involved much in the growth difference between the inner and outer rows of lettuce due to the nonhomogeneity of the temperature distribution inside the film tunnels. Once-over harvest of lettuce head is applied before January (Fig. 1a) and after that selective harvest is conducted by cutting two inner rows first (Fig. 1b and c). It takes about a week to do the second harvest of the outer rows. After the lettuces are selected and cut using a sharp cutting implement such as a knife, they are trimmed with a certain number of wrapper leaves left. Lettuce in Shizuoka is shipped using two packing techniques: naked pack or film wrap. Field packaged and shipped lettuce is cut at about the middle leaf, and trimmed to leave three wrapper leaves, then packed in the carton or plastic containers. Each container contains two layers of 12e18 heads depending on the head size. The top layer packed stump-up is washed to dissipate latex sap which excluded from the stump cut before shipping. For film-wrapped lettuce, heads are cut under the last basal leaf, trimmed to remain three to five wrapper leaves, then packed in containers and transported to storage facilities. The produces are then trimmed again to remain one wrapper leaf, dissipated the latex sap and wrapped with cellophane at the facilities before shipping. After harvest, the beds with plastic mulch will be used for further planting of sweet corn or even for the second transplantation of lettuce. 2.2. Design concept of the experimental harvester Generally, manually harvested lettuce is a mature, undamaged and unblemished head having the stump cut straight across to remain a desire number of wrapper leaves. Because winter head lettuce in Japan is grown inside plastic film tunnels, selective mechanical lettuce harvest is technically unfeasible. Some practices such as utilizing starter dressing or green house culture have been studied at the Shizuoka Prefectural Research Institute of Agriculture and Forestry to improve the growth uniformity of winter lettuce that can minimize the harvest lost due to immature heads if once-over harvest is applied. Consequently, the experimental lettuce harvester must meet the following requirements: (a) to mechanize once-over harvest of multi-row winter lettuce with efficiency four times higher than hand cut; (b) to injure and blemish the heads as less as possible; (c) not to damage the mulch; (d) to improve the working posture of workers; (e) for naked pack technique, lettuce heads will be trimmed to remove unwanted wrapper leaves, packed and dissipated the latex sap at the packing station;
Fig. 1. Conventional harvesting practices of head lettuce in Shizuoka prefecture.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
(f) for plastic film wrap technique, lettuce heads will be trimmed to remove unwanted wrapper leaves and packed. (g) harvested lettuce will be delivered out of field using a vegetable carrier. 2.3. Development of cutting component 2.3.1. Lettuce cutting methods Cutting unit is the most important component of a lettuce harvester and many previous studies have paid attention on design and development of a high accuracy cutting method for head lettuce. Suzuki et al. (2003) used a stationary knife to cut lettuce plant grown in one row per bed system. The knife was positioned near the rear end of front end tilted and angled gage discs which was connected to a parallel link. This cutting unit could cut lettuce stump at maximum harvester speed of 0.27 m/s without injuring lettuce heads. Researchers at The Institute of Agricultural Machinery (IAM-BRAIN, 2003) also developed an experimental lettuce harvester having similar gaging system for a powered arc-shaped cutting knife. The cutting unit could cut one row per bed lettuce at harvester speed of 0.06 m/s without cut-into heads and yield acceptable range of wrapper leaves. However, the cutting accuracy decreased as working speed increased. The rate of cut-into heads was approximately 20% at speed of 0.16 m/s. There have been several methods for cutting lettuce head in a two rows per bed system. Shepardson et al. (1974) developed a harvester for lettuce which was seeded two rows per bed. The author investigated the field performance of some cutting devices using stationary knife, band saw, blade saw, overlapping discs, and pinch-off discs. The front end height of the cutter heads was controlled by powered orientating discs. However, no devices were found accurately cut the stump of lettuce to ideally leave four outer leaves. Heads needed to be recut to obtain the quality of cut. Furthermore, dirt problem also occurred because low-positioned orientating discs scooped up some of the muck soil and the band saw or pinch-off discs scrubbed some soil into the stump cut. An investigation was made to determine if the 2-row lettuce grown on a specially flat bed could be cut accurately by a stationary knife (Adrian et al., 1976). The knife was adjustably mounted on a gage wheel that ran on the center of the bed. The height variation
3
Table 1 Specifications of cutting blade. Item
Specification
Material Hardness Dimensions (width thickness) Edge shape Sharpening angle Weight with sliding frame
Stainless steel (SUS-MS) HRC58 ± 2 1380 60 3.2 mm Smooth, single bevel 11 4260 g
between the row lines and the center of the bed was 2.5 cm. Result of cutting experiment indicated that cutting accuracy was greatest expectable with only 5% of cut-into heads. However, accuracy of this cutting method would depend on changes in cultural practices so substantial as to be economically unfeasible. Authors suggested that a reference rather than the bed center should be used. Water jet was also considered as a lettuce cutting method with advantages over other mechanical cutting methods such as minimization of acceleration and inertia problems and optimization of space requirement for cutting mechanism (Schield et al., 1973). Field test result showed that lettuce stumps could be cut with a water jet but no data on the cutting accuracy was reported. Furthermore, problem of controlling the nozzle path occurred as the harvesting module advanced down the row. The method also had potential problems such as unclean cut, dirt, high power requirement, water supply and dangerous high-pressure nozzle. 2.3.2. Design of cutting component Based on the available cutting methods together with the lettuce cultural system in Shizuoka, we have been investigating the cutting accuracy of a cutting component (Fig. 2) that uses a reciprocating blade to slice lettuce stems at a predetermined cutting height of the blade with respect to soil bed surface. This cutting method was chosen because it can simultaneously cut the four rows of lettuce grown on a fairly flat bed system, which is used in lettuce production areas in Shizuoka (Nang et al., 2013). In addition, the coherent problems of mechanical cutting mechanism, which are high acceleration and inertia, heavy weight, large space requirement for positioning blades, actuating devices and linkages,
Fig. 2. Schematic view of reciprocating-blade cutting component.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
4
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
can be eliminated by optimal design for weight and dimension of the cutting component. The structure of reciprocating-blade cutting component was described in details in a previous report (Nang et al., 2013). In this development, the cutting bade that had a smooth single beveled edge was made of 3.2 mm thick stainless steel plate and sharpened at an angle of 11. Its' detailed specifications are reported in Table 1. The effects of blade thickness, edge shape (scalloped or serrated) and sharpening angle on the cutting quality and torque requirement will be taken into account in a future study. The cutting component is positioned at a front portion of the harvester and adjustably mounted on a frame of the lifting component. The cutting and lifting component rides on powered gage wheel having five rollers that are positioned behind the cutting blade and between the lettuce rows to follow the bed contour. The gap between rollers is large enough so that the gage wheel does not run over the cut stump left on the bed surface. The rear end of lifting component frame is mounted on harvester main frame by a universal joint that let the gage wheel follows the bed contour even the left and right furrows have different depth. 2.3.3. Lettuce stem cutting experiments with reciprocating bade Cutting experiments were conducted to verify performance of reciprocating-blade cutting component at different combinations of reciprocating stroke, frequencies and harvesting speed. In the previous study (Nang et al., 2013), the measurements of lateral cutting resistance, required cutting torque and power for slicing two lettuce stems using a stationary blade were made at reciprocating frequency and stroke of 4 Hz and 18 mm, respectively and at different forward speeds. In this report, power requirement for cutting two and four lettuce stumps was also measured as a reference for choosing driving motor of cutting blade at different reciprocating frequencies. Experimental apparatus is depicted in Fig. 3. A movable carrier that simulated the soil bed of lettuce carried two or four stumps (average stem diameter of 40 ± 5 mm) of over-mature lettuce plants through the stationary cutting component at forward speed of 0.1 m/s. The cutting blade was reciprocated at a stroke of 18 mm, and rotational speeds of 120, 240, and 360 rpm or blade reciprocating frequencies of 2, 4, and 6 Hz, respectively. Torque for driving cutting blade together with rotational speed of crank shaft was measured using a torque meter (TQR-300N,
Nikkei), an encoder (E6C2-CWZ6C, Omron). Crank shaft angle was detected by a proximity sensor (TL-W5E1 5M, Omron). The measured data were recorded through a 10 Hz low-pass filter at a sampling rate of 1 KHz using a data acquisition system (Memory Hicorder 8842, Hioki). Each experiment was replicated 5 times. 2.4. Development of lifting component 2.4.1. Methods for lifting cut lettuce heads from ground Lifting component takes cut produce from the back of cutting component and carries them onto the horizontal packing conveyor without injuring the heads or remained wrapper leaves. Shepardson et al. (1974) used rod-type chain belting to elevate cut lettuce and eliminate dirt at an inclined angle below 28 . This component could not convey heads in any given orientation and it delivered debris along with lettuce. The authors also used two side gripping resilient belts to receive cut lettuce and maintain its orientation for transfer of the produce to the elevating conveyor consisting of top and bottom belts. Excess cut off leaves, undersized heads and debris dropped free from this conveyor. Rubber-fingered belts (Lenker et al., 1973), fan-shaped sponge conveyors (Suzuki et al., 2003), and wave-formed belts (IAM-BRAIN, 2003) were used to grasp lettuce by their sides just before they are cut and then lift the cut heads to the trimming conveyor. The introduced methods for lifting cut produce onto trimming platform were applied for one row per bed system or two rows per bed system with large enough row spacing for positioning lifting belts. 2.4.2. Design of lifting component Because Shizuoka winter lettuce is grown four rows per bed with condensed zigzag pattern, the principle of lifting cut heads by their left and right side using two-side gripping belts appeared unfeasible. Therefore a top and bottom belts were used to elevate lettuce to the packing conveyor (Fig. 4). The bottom belt that travels upward is 1100 cm wide and is made of urethane and polyester. The top belt is a rod-type chain belting system, which consist of four wave-formed urethane belts fixed to the rods at a pitch of 20 cm. The wave-formed part is 150 cm wide, 20 cm high and positioned in zigzag pitch pattern. The arrangement of wave-formed parts significantly affects the performance of lifting component so their optimal size will be considered in the next step of machine development. The top belt travels downward and resiliently
Fig. 3. Experimental apparatus for cutting lettuce stems.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
5
Fig. 4. Schematic view lifting component mounted on the experimental lettuce harvester.
compresses the heads as they are cut then together with the bottom belt gasp and lift the produce to the packing station. The contact area between lettuce head and the wave-formed belts are covered with 5 mm-thick sponge sheets to reduce impact force of the belts on the heads as well as to increase friction between them hence improve the lifting performance of the top belt. The lifting component conveys cut heads at an inclined angle that varies from 28 to 33 depending on bed height. During harvesting operation, the front end of lifting component rides on the five-roller gage wheel. In travelling condition, the cutting and lifting components are raised up by two arms that are actuated by an electric powered hydraulic cylinder. The experimental harvester is equipped with 2 electric generators with total power of 2.5 kW and with rubber tracks, which allow a regular forward movement on any kind of terrains. The track-center distance can be varied from 1.6 to 2 m. In the final step of development, the harvester will be equipped with an automatic pilot apparatus so that only a packing crew is on board during harvesting operation. 2.5. Field performance of developed harvester Field tests were conducted to evaluate the performances of reciprocating-blade cutting component and lifting component. Winter lettuce was grown on fairly flat beds having width of 2 m and height of 20 cm in an experimental field of Shizuoka Prefectural Research Institute of Agriculture and Forestry (Fig. 5). Growth conditions of lettuce before harvesting are reported in Table 2. The
tests were performed at harvesting speeds up to 0.24 m/s with cutting stroke of 18 mm and blade reciprocating of 4 Hz. Although the optimal cutting height to remain 2 to 3 wrapper leaves on the cut head was approximately 16 mm, initial height of cutting blade was set 10 mm above bed level for every test in order to minimize the number of heads with too few wrapper leaves. Two packing workers walked behind the harvester on the bed surface to pack cut heads into plastic containers. Working functions of cutting and lifting components were investigated for every test while quality of lettuce in terms of remained wrapper leaves, damaged rate, dirt problem were measured to evaluate the accuracy of cutting component as well as the delivery performance of the lifting component. 3. Results and discussions 3.1. Results of cutting experiments The results of cutting experiments indicated that the reciprocating cutting component could smoothly cut lettuce stumps. Sample data of measured torques for cutting lettuce stumps at different blade reciprocating frequencies are displayed in Fig. 6. The measured total torque was a sum of unloading torque for reciprocating the cutting blade in noncutting condition and torque for cutting lettuce stem. The unloading torque was synchronized with the total cutting torque with respect to the crank angle. A sample of synchronized data of one rotation of the crank shaft in which the maximum total torque occurred is depicted in Fig. 7. The unloading torque had a wave form as a function of crank angle with fluctuation amplitude. It changed direction 3 times in each rotation of the crank shaft and fluctuation pattern was similar for both the forward and return strokes. The torque increased in the
Table 2 Physical characteristics of harvested lettuce.
Fig. 5. Overview of developed experimental harvester.
Variety Total plant weight, g Total plant height, cm Number of wrapper leaves Head weight, g Head height, cm Head transverse diameter, cm Head conjugate diameter, cm Optimal cutting height, mm
M-WRAP-231 599 ± 128 18 ± 1 12 ± 2 396 ± 64 16 ± 1 16 ± 2 14 ± 2 16 ± 5
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
Measured total torque (N.m)
6
1.2 Reciprocating frequency 1
2Hz
4Hz
6Hz
0.8 0.6 0.4 0.2
0 -0.2 0
0.5
1
1.5
2
2.5
3
Time (s) Fig. 6. A sample of measured toque for cutting four lettuce stumps.
first half of both forward and return strokes and then decreased at the second halves. It reached maximum and minimum values at the return stroke. This was contributed to the acceleration and inertia of the cutting blade. The mean value of maximum unloading torques increased with increasing rotational speed of crank shaft and was 0.16, 0.35 and 0.49 Nm for reciprocating frequencies of 2, 4 and 6 Hz, respectively (Fig. 8). The torque measured during lettuce stump cutting was greater than that in unloading condition and reached a peak several times before the lettuce stems were totally cut. Theoretically, the amplitude of the total torque depends on the reciprocating mass, the cutting resistance and the moment that the cutting blade started slicing the lettuce stems. However, the maximum value of the total torque always occurred at the end of the forward stroke. The mean value of maximum total torques was proportional to rotational speed of crank shaft. In the case of 2-stump cutting, it was 0.35, 0.56 and 0.74 Nm corresponding to reciprocating frequencies of 2, 4 and 6 Hz, respectively, while that was 0.70, 0.94 and 1.06 Nm for cutting 4 stumps, almost twice greater than cutting 2 stumps (Fig. 8). Meanwhile, the torques for cutting 2 or 4 stumps were slightly increased with increasing reciprocating frequency. The mean of maximum values of cutting torques was 0.28, 0.31, 0.32 Nm in the case of 2-stump cutting and 0.60, 0.67, 0.72 Nm in the case of 4stump cutting with respect to reciprocating frequencies of 2, 4 and 6 Hz, respectively. The required power for cutting lettuce stumps was calculated from total torque and rotational speed of crank shaft. It was 4.3, 14.0, 27.7 W for cutting 2 lettuce stumps and 8.9, 23.6, 39.8 W for cutting 4 lettuce stumps corresponding to reciprocating frequencies of 2, 4 and 6 Hz, respectively (Fig. 8).
3.2. Results of field tests Field test results indicated that the experimental harvester could harvest lettuce at ground speed up to 0.24 m/s. Satisfactory operation with packing crew of 2 workers only were obtained at harvesting speeds under 0.10 m/s. For higher speed, more packing workers or bulk packing of harvested lettuce would be necessary. The cutting component properly functioned at reciprocating stroke of 18 mm and frequency of 4 Hz and successfully cut lettuce plants (Fig. 9). At these reciprocating parameter, vibration of the component was small so that it is unnecessary to reduce the acceleration and inertia of the cutting mechanism by using counterweight for crank shaft. Generally, cutting blade reciprocates just above the mulch so dirt problem was not observed and clean cuts of lettuce stumps were also obtained. The gage wheel could follow the bed contour but it was not capable of controlling the height of cutting blade on irregular bed surface because the distance from front end of the blade to the axle center of the gage wheel is rather long, approximately 30 cm. The diameter of the gage wheel should be reduced to shorten this distance in order to improve the ability of cutting height control. The cutting blade should be positioned near the front end of the bottom belt of the lifting component to facilitate the cut head delivery. The lifting component could also successfully elevate the cut heads to the packing belt (Fig. 10). The font end of bottom belt should be position near the bed surface to smooth the head delivery. Some bruises or breakage of wrapper leaves were occurred as the heads were elevated from the cutting blade to the bottom belt. However the heads were salvaged by further trimming of unwanted wrapper leaves. The traveling speed of the top and
Fig. 7. A sample of synchronized toque for cutting four lettuce stumps.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
Total torque Cutting torque Power requirement
1.2
90
0.8
60
0.4
30
0.0
0 0 2 4 6 8 Reciprocating frequency (Hz)
120
1.6 Cutting 4 stumps
Torque (Nm)
Torque (Nm)
Cutting 2 stumps
1.2
90
0.8
60
0.4
30
0.0
0 0 2 4 6 8 Reciprocating frequency (Hz)
Power requirement (W)
120
Power requirement (W)
1.6
Unloading torque Total torque Cutting torque Power requirement
7
Fig. 8. Measured and computed torque for different reciprocating frequency.
Fig. 9. Lettuce stumps left on the bed (left) and harvested heads (right).
bottom belts were synchronized with harvesting speed to minimize the head and wrapper leaf injures due to relative motion between lettuce and lifting component. The stumps set on the bottom belt and the head was held by the resilient top belt. The wrapper leaves. The component conveyed mature heads along with cut off leaves, undersized heads, undeveloped heads, and debris. Only commercialized heads were sorted on the packing conveyor and packed into plastic containers, while other was dropped freely on to the bed surface. Investigations of machine harvested lettuce quality were conducted for harvesting speeds of 0.04 and 0.08 m/s. The results are reported in Table 3. Because the pre-set of blade cutting height was lower than optimal cutting height, the percentage of acceptable
heads with 2e6 wrapper leaves was only 46.8 for harvesting speed of 0.04 m/s. However that increased to 73.1% at higher speed of 0.08 m/s. This indicated that the pre-set cutting height could not assure a desirable cutting accuracy with acceptable number of wrapper leaves. An active method of controlling the cutting height of blade during harvesting operation is required. The percentage of heads with less than 2 wrapper leaves, including the cut-into heads, was 2.9 and 10.6% for harvesting speeds of 0.04 and 0.08 m/s, respectively. This was partly due to variations between lettuce row line height, and partly due to the variations in length of lettuce stumps which resulted from nonuniformity of plant growth. There were some harvested heads with ribs of wrapper leaves that were bruised and/or broken at the gap between the cutting blade and the front end of the bottom conveyor or somewhere along the lifting component frame. However these heads were salvaged by the trimming cut. The head injure was due to the interaction between the top waveformed belts and the delivered heads that occurred when there was a difference in speed of the harvester and the lifting component. The rate of head injure increased with harvesting
Table 3 Quality of machine harvested lettuce. Commercialized Ground Number of Cut quality (%) head ratea speed investigated Too few Acceptable Too many Head heads (m/s) (%) leaves leaves injure 0.04 0.08
Fig. 10. Side view of lifting component in operation.
200 192
2.9 10.6
46.8 73.1
50.3 16.3
2.8 7.2
94.5 87.2
a Commercialized head rate: percentage of machine-harvested heads having greater or equal to two wrapper leaves without head injuries. Note that some injured cut heads would also have less than two wrapper leaves or vice versus.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002
8
V.N. Nang, S. Yamane / Engineering in Agriculture, Environment and Food xxx (2014) 1e8
speed and was 2.8 and 7.2% for harvesting speeds of 0.04 and 0.08 m/s, respectively. The heads with number of wrapper leaves less than 2 or with head injuries were classified as unmarketable heads. Consequently, with a further trimming, the rate of commercialized head was 94.5 and 87.2% corresponding harvesting speeds of 0.04 and 0.08 m/s. Improvements for more accurate performance of cutting and lifting components are needed to obtain the targeted commercialized of harvested heads of over 95%.
2. Cut lettuce can be successfully elevated to the packing belt by the developed lifting component consisting of a bottom and top belts. 3. Maximum torque and power requirement for cutting 2 lettuce plants were 0.73 and 27.7 Nm, respectively. 4. For harvesting speed under 0.10 m/s, the experimental harvester could harvest lettuce with rate of commercialized heads ranging from 87.2 to 94.5%. 5. Accuracy of lettuce cutting of the experimental harvester was affected by the uniformity of soil bed profile and lettuce growth.
4. Conclusions
References
This study reports the development of a prototype harvester for winter-clipping head lettuce in Japan. The research and development of the individual components assembled in to an experimental lettuce harvester indicated the feasibility of mechanical harvesting of head lettuce. The prototype machine that was implemented with a reciprocating-blade cutting component could harvest lettuce with almost acceptable marketable rate at forward speed less than 0.1 m/s. Some conclusions could be drawn from the experimental work as following. 1. Lettuce stumps can be successfully cut by the developed cutting component with a reciprocating blade.
Adrian PA, Lenker DH, Nascimento D. Concepts and development of an experimental mechanical trimmer for crisphead lettuce. Trans ASAE 1976;19(1): 17e20. Institute of Agricultural Machine, Bio-oriented Technology Research Advancement Institution (IAM-BRAIN). Development of a Crisphead lettuce harvester. Annual Technical Research Report. 2003. p. 136e7. Nang V.N., Suguru Y., Cutting component development for a lettuce harvester, Annual International Meeting of American Society of Agricultural and Biological Engineers, Kansas City, Missouri, July 21eJuly 24, 2013, Paper No. 1316326132013. Schield M, Harriott BL. Cutting lettuce stems with a water jet. Trans ASAE 1973;16(2):440e2. Shepardson ES, Pollock JG, Rehkugler GE. Research and development of a lettuce harvester. Trans ASAE 1974;17(2):212e6. Suzuki H, K.Saito K, Shiokawa H, Sato K, Okada K, Mizutani K, et al. Lettuce harvester development (Part 4) e development of trial machine for practical use and field examinations. J Jpn Soc Agric Mach 2003;65(6):141e8.
Please cite this article in press as: Nang VN, Yamane S, Development of prototype harvester for head lettuce, Engineering in Agriculture, Environment and Food (2014), http://dx.doi.org/10.1016/j.eaef.2014.09.002