Performance Comparison and Evaluation of Two Small Chili Pepper Harvester Prototypes That Attach to Walking Cultivators

Abstract

Two prototypes of small chili pepper harvesters that attach to walking cultivators were designed and manufactured before field performance tests were conducted. The two prototypes were designed and manufactured with different main frame materials, forms of divider, picking guides, and helix rotation speeds. The maximum helix speed of the first prototype was 500 rpm, and the helix rotation speeds of the second prototype were a minimum of 510 rpm and a maximum of 730 rpm. Field performance tests were conducted on two species of chili, the AR Legend and the Jeokyoung, to determine which was suitable for mechanization. The Jeokyoung species was found to be most suitable for mechanization as its harvest efficiency was higher and its pepper left on plant rate and ground fall loss rate were lower than AR Legend’s. When the first and second prototypes were compared at helix rotation speeds of 500 to 510 rpm, in the case of the AR Legend, the average harvest efficiency of the second prototype was higher than the first prototype by 2.2%, the average pepper left on plant rate was lower by 2.1%, and the average ground fall loss rate was lower by 3.9%. In the case of the Jeokyoung, the performance of the second prototype was further improved over the first prototype as the average harvest efficiency increased to 5.2%, and the difference in average ground fall loss rate increased to 8.8%.

1. Introduction

Peppers account for 22% of condiment vegetables produced in Korea, and they are used to produce the main condiment favored by Korean cuisine due to their unique spicy flavor. In Korea, pepper cultivation area and production weight are reported to have decreased by more than 40% and 30%, respectively, from 53,097 ha and 352,977 tons in 2006 to 28,337 ha and 241,946 tons in 2017 due to a decrease in the agricultural population and the low mechanization rate compared to other field-based crops [1]. Even though the mechanization rates of tillage and soil preparation, mulching, and pest control are 97.7%, 58.5%, and 90.9%, respectively, pepper harvesting has not been mechanized at all due to the growth characteristics of peppers that mean they need to be continuously harvested four to six times (or more) over the course of a season [2,3]. The National Institute of Horticultural and Herbal Science (Jeonju, Jeollabuk-do, Korea) has developed the Jeokyoung, a chili species suited to mechanical harvesting, so a pepper harvesting machine suitable for this species needs to be developed. Nam et al. [4] analyzed the mechanical and physical characteristics of Korean pepper species to find a species suitable for use with a pepper harvesting machine. In their analysis, the Poisson’s ratio, modulus of elasticity, shear modulus, density, recovery factor, and the friction coefficient of the “AR Legend” species were investigated; the Poisson’s ratio, modulus of elasticity, and the shear modulus were found to be 0.295, 3.295 × 107 Pa, and 4.624 × 106 Pa, respectively.

Marshall and Boese [5] proposed and developed a helical type picking design for mechanical pepper harvesting; using this design the yield of pepper was found to increase as the rotation speed of the helical shaft increased. Choi [6] produced picking designs in the form of three-row helical, four-row helical, bar, and semi-circular tooth designs that were evaluated in various tests. The bar type was found to be the picking tooth type most suitable for pepper harvest showing a successfully harvested rate of 77.3% and a pepper left on plant rate of 3%. Paul and Walker [7] conducted a picking performance test for five types of picking design (disk, chain, Creager, Hernandez, and helix) to find the best type. Among these five types, the helix type was found to be the best type for picking, showing a harvest efficiency of 88.3% and a mechanical damage rate of 11.2%. Studies were conducted recently in Korea to develop a self-propelled pepper harvester by improving the existing picking and selecting designs [8,9,10,11,12,13]. Field performance tests were conducted for picking with different helix types to determine an ideal design for picking; in these tests, the maximum harvest efficiency was found to be 76%, and the lowest ground fall loss rate and the pepper left on plant rate were found to be 15% and 7%, respectively, for the triple helix design with a shaft speed of 300 rpm, driving speed of 0.3 m/s, and a picking part inclination angle of 40° [12,13]. Nam et al. [9] attempted to improve the harvest efficiency by additionally designing and producing a second picking design using a drum. Jo et al. [10] designed and produced a sorting factor test device of a card cleaner type. They conducted a performance test looking at picking out peppers from foreign substances (stone, pepper twig, etc.) using their design. The conditions that led to the fewest foreign substances being picked were an inclination angle of 15° and rotation speed of 50 rpm. Byum et al. [11] determined the ideal conditions for a card cleaner using a computer simulation with the discrete element method (EDEM). They compared the results with the factor performance test conducted by Jo et al. [10]. The self-propelled pepper harvester developed using the chassis base of a combine [9,10,12,13] is suitable for use in large pepper farms not smaller than 1 ha. However, this kind of large chili pepper harvester is not ideal because farms smaller than 1 ha account for 65.7% of the total pepper farms based on 2017 data. Motivated by this, the need for a small pepper harvester attached to a walking cultivator is clear.

The objective of the present study was to design and produce two prototype small chili pepper harvesters that attach to walking cultivators and conduct field performance tests on them. The harvest efficiency, successfully harvested rate, pepper with twig rate, pepper left on plant rate, ground fall loss rate, and the damage rate were investigated, and the two prototypes were comparatively analyzed using the results. The harvest performance was also comparatively analyzed depending on the pepper species and the rotation speed of the helical shaft.

2. Materials and Methods

2.1. Small Chili Pepper Harvesters Attached to Walking Cultivators

The small chili pepper harvesters attached to walking cultivators (PHAWCs) have picking parts attached to a walking cultivator, and the power take-off (PTO) power of the walking cultivator is used. The first PHAWC prototype comprises a main frame that supports the main body, a joint that couples the picking part with the walking cultivator, a picking part that picks the pepper fruit, and a walking cultivator that generates power, as shown in Figure 1. Its size is 2530 × 830 × 1280 (L × W × H) mm, and the material of the main frame is carbon steel pipe (STPG). The weight of the picking part, main frame, and the joint, excluding the walking cultivator, is 178 kg; the total weight including the walking cultivator, which weighs 112 kg, is 289 kg, making it difficult to operate or handle. In addition, there is the problem that a bending moment may be generated by its own weight at the joint that connects the main frame and the walking cultivator. Accordingly, the weight of the main frame needs to be lightened for simple operation and increased durability. The second PHAWC prototype was produced using aluminum as the main frame material. The weight of the main frame, joint, and the picking part was reduced to 106 kg, lighter than the first prototype by 40.5%. This was partially achieved by reducing the gear box size of the second PHAWC prototype by 51.4% from that of the first prototype. The design drawing of the second PHAWC prototype and the appearance of the product are shown in Figure 2. It comprises a main frame, joint, picking part, and a walking cultivator in the same way as the first PHAWC prototype. The second PHAWC prototype was produced at a size of 1446 × 720 × 1300 (L × W × H) mm, and the total weight including the walking cultivator weight of 155 kg was 261 kg.

2.1.1. Walking Cultivator

As the duration of the contact between the helix and pepper fruit is determined by the walking speed of the pepper harvester, it is an important factor related to the harvest efficiency. To secure high harvest efficiency, the walking speed of the walking cultivator is required to be 0.30 m/s or slower. Accordingly, among the walking cultivators sold in Korea, the AMC-900SM (Asia Technology Co., Daegu, Daegu, Korea) was chosen for the first PHAWC prototype due to the minimum walking speed in first gear of the transmission being 0.28 m/s. For the second prototype, the walking cultivator KMC-750 (Tongyang Moolsan Co., Seoul, Seoul, Korea) was chosen due to the minimum possible working speed in first gear of the transmission being 0.25 m/s. The specifications of the two prototypes are shown in

Table 1. Specifications of cultivators.

Item	AMC-900SM	KMC-750
Weight (kg)	112	155
Rated power (kW)	6.7	7.4
Power take-off speed (rpm)	714–1350	650–1300
Working speed at first gear (m/s)	0.28	0.25
Rotational speed of helix when attaching each prototype (rpm)	400–500	510–730

below:

2.1.2. Production of Helix’s Helical Shaft and Setting of Rotation Speed

The helix that makes up the picking part performs the task of picking pepper fruit and was designed to have the form of a double helix where two helical bars are wound to have an angle of 180° around the rotating shaft (Figure 3). The most important factors in producing a helix are helix type, pitch, length, wire diameter, inner diameter, shaft diameter, and the distance between helixes [14]. The helix length was downsized to 900 mm to suit the purpose based on the results found in the picking part factor test [15]. The shape and data of the helix are shown in Figure 4 and

Table 2. Specifications of helix.

Helix Factor	Parameter
Type of helix	Double helix
Pitch of helix	450 mm
Length of helix	900 mm
Wire diameter of helix	15 mm
Inner diameter of helix	80 mm
Diameter of Shaft	34 mm
Distance between two helixes	16 mm

. Kim [15] reported that the rotation speed of helical shaft is the factor that has the biggest effect on the harvest efficiency. Depending on the rotation speed of the helical shaft, the first PHAWC prototype can be operated at speeds ranging from a minimum of 400 rpm to a maximum of 500 rpm, and the second PHAWC prototype can be operated at speeds ranging from a minimum of 510 rpm to a maximum of 730 rpm (Table 1).

2.1.3. Shapes of Divider and Picking Guide

The divider was installed in front of the helix to enable peppers to be stably brought into the helix; the divider of the second PHAWC prototype was produced deliberately in the form similar to the divider of a combine to facilitate inflow of peppers (Figure 5). The picking guide is a device that helps to make it possible to pick whole pepper plants that are longer than the helix. It is installed at the top of the helix and makes it possible to pick even the pepper fruits at the top of the plant by pressing down on the top of the pepper plants when they move along the conveying part of the helix. The dividers were designed to enable entire pepper plants to be picked up by working in conjunction with the picking guides of the first and second PHAWC prototypes that come in the form of a bar at the center or a wedge, respectively (Figure 6).

2.2. Test Sample

In the present study, a field test was conducted to evaluate the performance of the two PHAWC prototypes. The AR Legend and the Jeokyoung were selected to be suitable for mechanization and were used as the test samples (Figure 7). The Jeokyoung is a species fostered in the National Institute of Horticultural and Herbal Science. It is a species that has a strong spicy taste, is pigment-rich, and has a high single harvest rate and ripe fruit yield. The AR Legend is a species fostered by Company P and is characterized by a strong resistance to the anthrax disease. The pepper samples used for the performance evaluation were cultivated outdoors in National Institute of Horticultural and Herbal Science of Rural Development Administration, and the evaluation was carried out 104 days after sowing.

2.3. Field Test Method

To evaluate the performance of the first and second PHAWC prototypes, a field performance test was conducted using two species of pepper (AR Legend and Jeokyoung). In the case of the first PHAWC prototype, as no pepper was harvested at the helix rotation speed of 400 to 450 rpm, it was operated at the maximum helix rotation speed of 500 rpm. In the case of the second PHAWC prototype, the test was conducted by setting the helix rotation speed to three levels: 510, 610, and 730 rpm. Pepper plants in five furrows were used for each test, and the test was repeated three times. After harvesting, the harvested peppers, peppers that had fallen on the ground, damaged peppers, and foreign substances gathered in each section were collected and weighed (Figure 8).

2.4. Analysis Method

Choi [6] measured the weight of the harvested pepper fruits, weight of the pepper fruits harvested with no twigs, weight of the pepper fruits harvested with twigs, weight of the pepper fruits left on the pepper plants, weight of the pepper fruits that had fallen on the ground, and the weight of the damaged pepper fruits and used them to investigate the harvest efficiency, successfully harvested rate, pepper with twig rate, pepper left on plant rate, ground fall loss rate, and the damage rate. Harvest efficiency is the weight percentage of the harvested pepper fruits to all pepper fruits (1), and the successfully harvested rate is the weight percentage of the pepper fruits harvested with no twigs to all pepper fruits (2). The pepper with twig rate is the weight percentage of the pepper fruits harvested with twigs to all pepper fruits (3), and the pepper left on plant rate is the weight percentage of the pepper fruits left on the pepper plants to all fruits (4). The ground fall loss rate is the weight percentage of the pepper fruits that have fallen on the ground to all fruits (5), and the damage rate is the weight percentage of the damaged pepper fruits to all pepper fruits (6).

$$D_{he}=\frac{P_{hp}}{P_{ap}}\times 100$$

(1)

where $D_{he}$ = harvest Efficiency, %, $P_{hp}$ = weight of harvested pepper fruits, g, $P_{ap}$ = weight of all pepper fruits, g

$$D_{shr}=\frac{P_{ph}}{P_{ap}}\times 100$$

(2)

where $D_{shr}$ = successfully harvested rate, %, $P_{ph}$ = weight of pepper fruits harvested with no twigs, g, $P_{ap}$ = weight of all pepper fruits, g

$$D_{ptr}=\frac{P_{pht}}{P_{ap}}\times 100$$

(3)

where $D_{ptr}$ = pepper with twig rate, %, $P_{pht}$ = weight of pepper fruits harvested with twigs, g, $P_{ap}$ = weight of all pepper fruits, g

$$D_{plr}=\frac{P_{pl}}{P_{ap}}\times 100$$

(4)

where $D_{plr}$ = pepper left on plant rate, %, $P_{pl}$ = weight of pepper fruit left on the pepper plants, g, $P_{ap}$ = weight of all pepper fruits, g

$$D_{glr}=\frac{P_{pfg}}{P_{tp}}\times 100$$

(5)

where $D_{glr}$ = ground fall loss rate, %, $P_{pfg}$ = weight of pepper fruits that have fallen on the ground, g, $P_{ap}$ = weight of all pepper fruits, g

$$D_{dr}=\frac{P_{dp}}{P_{tp}}\times 100$$

(6)

where $D_{dr}$ = damage rate, %, $P_{dp}$ = weight of damaged pepper fruits, g, $P_{ap}$ = weight of all pepper fruits, g

3. Results and Discussion

The results of the field performance tests conducted for the first PHAWC prototype are as shown in

Table 3. Experiment results of field test based on rotational speed of helix and variety of pepper for each prototype.

Prototype	Variety	Rotational Speed of Helix (RPM)	AVE/SD	HE 1 (%)	SHR 2 (%)	PTR 3 (%)	PLR 4 (%)	GLR 5 (%)	DR 6 (%)
First version	AR	500	AVE 7	75.8	63.9	18.0	11.0	16.9	-
	Legend		SD 8	5.6	6.2	7.8	3.2	2.1	-
	Jeokyoung	500	AVE	81.9	64.7	17.3	5.7	12.3	-
			SD	4.6	12.3	12.9	2.3	2.7	-
Second version		510	AVE	78.0	46.2	31.2	8.9	13.0	0.7
			SD	8.1	5.5	7.0	2.2	7.4	0.6
	AR	610	AVE	86.2	46.2	38.1	7.0	6.9	2.3
	Legend		SD	1.6	8.8	8.4	2.1	2.7	1.2
		730	AVE	91.7	45.6	44.3	1.9	6.3	2.3
			SD	5.7	5.4	11.1	0.6	5.2	1.2
		510	AVE	87.2	64.0	21.4	8.0	3.5	2.0
			SD	0.9	5.6	4.9	3.6	1.8	1.0
	Jeokyoung	610	AVE	93.8	63.5	28.2	2.8	3.5	2.0
			SD	3.2	15.5	17.1	2.2	1.7	2.0
		730	AVE	95.3	70.1	23.1	1.2	3.5	2.3
			SD	2.7	4.4	5.2	0.5	2.7	0.6

¹ HE: harvest efficiency; ² SHR: successfully harvested rate; ³ PTR: pepper with twig rate; ⁴ PLR: pepper left on plant rate; ⁵ GLR: ground fall loss rate; ⁶ DR: damage rate; ⁷ AVE: average; ⁸ SD: standard deviation.

. The average harvest efficiency is shown to be 75.8% for AR Legend and 85.6% for Jeokyoung. It shows an average successfully harvested rate of about 64% and an average pepper with twig rate of about 17% to 18% for the two species. It shows average pepper left on plant rates of 5.7% to 11.0% and average ground fall loss rates of 12.3% to 16.9%, while no damage occurred at the helix rotation speed of 500 rpm in the case of the first PHAWC prototype.

The results of the field performance test conducted for the second PHAWC prototype are shown in Table 3. The result shows average harvest efficiency of 78.0% to 95.3%, average successfully harvested rates of 45.6% to 70.1%, average pepper with twig rates of 21.4% to 44.3%, average pepper left on plant rates of 1.2% to 8.9%, average ground fall loss rates of 3.5% to 13.0%, and average damage rate of 0.7% to 2.3%. When the helix rotation speed of the second PHAWC prototype increased from 510 to 730 rpm, the harvest efficiencies of both species increased, and the pepper left on plant rates and the ground fall loss rates decreased. In the case of the AR Legend, though the pepper with twig rate gradually increased (from 31.2% to 44.3%) as the helix rotation speed of the second PHAWC prototype increased (from 510 to 730 rpm), the successfully harvested rate did not appear to depend on the helix rotation speed. In the case of the Jeokyoung, though the successfully harvested rate gradually increased (from 64% to 70.1%) as the helix rotation speed of the second PHAWC prototype increased (from 510 to 730 rpm), the pepper with twig rate did not appear to depend on the helix rotation speed.

The Jeokyoung is thought to be a species most suitable for mechanization because its average harvest efficiency is higher, and the average pepper left on plant rate and the average ground fall loss rate are lower than those of the AR Legend when comparing the two species using the first and second PHAWC prototypes.

With the AR Legend, the average harvest efficiency of the second PHAWC prototype was greater than that of the first PHAWC prototype by a minimum of 2.2% (510 rpm) and a of maximum 15.9% (730 rpm), and the average successfully harvested rate of first PHAWC prototype was greater by about 18% (510, 610, 730 rpm) for the second PHAWC prototype. The average pepper with twig rate was better for the second PHAWC prototype by a minimum of 13.2% (510 rpm) and a maximum of 26.3% (730 rpm), while the average pepper left on plant rate was lower for the second PHAWC prototype by a minimum of 2.1% (510 rpm) and a maximum of 9.1% (730 rpm). The average ground fall loss rate was lower for the second PHAWC prototype by a minimum of 3.9% (510 rpm) and a maximum of 10.6% (730 rpm).

With the Jeokyoung, the average harvest efficiency of the second PHAWC prototype was better than the first PHAWC prototype by a minimum of 5.2% (510 rpm) and a maximum of 13.3% (730 rpm), and the average pepper with twig rate for the second PHAWC prototype was better by a minimum of 4.1% (510 rpm) and a maximum of 10.9% (610 rpm). The average ground fall loss rate for the second PHAWC prototype was lower by 8.8% at all helix rotation speeds (510 to 730 rpm). It can be seen that the performance of the second PHAWC prototype is improved in comparison to the first PHAWC prototype as the harvest efficiency, which is the most important factor in harvesting work, increased, and the pepper left on plant rate and the ground fall loss rate decreased; however, with the second prototype, additional work is required to separate fruits attached to twigs because its pepper with twig rate is greater than the first PHAWC prototype. Accordingly, additional study is required to reduce the pepper with twig rate of the second PHAWC prototype.

With the Jeokyoung, the average harvest efficiency of the second PHAWC prototype (87.2–95.3%) is higher than Choi [6] (77.3%) and Paul and Walker [7] (88.3%). The average damage rate of the second PHAWC prototype (2.0–2.3%) is also much lower than Paul and Walker [7] (11.2%).

The results of conducting a variance analysis using a two-way layout with replication to determine the effects of species and helix rotation speed on the harvest efficiency, successfully harvested rate, and pepper left on plant rate of the second PHAWC prototype are shown in

Table 4. Statistical results of field test based on rotational speed of helix and variety of pepper.

Item	Source	DF	ANOVA SS	Mean Square	F Value	Pr > F
HE 1	Variety	1	214.7282	214.7282	10.78	0.0065
	Rotational speed (rpm)	2	365.1086	182.5543	9.16	0.0038
	Variety*rpm	2	22.3976	11.1988	0.56	0.5843
	Error	12	239.0906	19.9242
	Total	17	841.3251
SHR 2	Variety	1	1780.0555	1780.0556	25.43	0.0003
	Rotational speed (rpm)	2	32.4444	16.2222	0.23	0.7966
	Variety*rpm	2	44.4444	22.2222	0.32	0.7339
	Error	12	840.0000	70.0000
	Total	17	8696.9444
PLR 3	Variety	1	17.0917	17.0917	3.78	0.0755
	Rotational speed (rpm)	2	142.9982	71.4991	15.83	0.0004
	Variety*rpm	2	11.0023	5.5011	1.22	0.3299
	Error	12	54.1964	4.5163
	Total	17	225.2888

¹ HE: harvest efficiency; ² SHR: successfully harvested rate; ³ PLR: pepper left on plant rate.

. For the variance analysis, SAS (Version 9.3, SAS Institute, North Carolina, Cary, USA, 2018) commercial software was used as the statistical program. The results of carrying out a variance analysis on the effect of the species or helix rotation speed on the pepper with twig rate, ground fall loss rate, and damage rate showed that there was no significant effect as all P-values were greater than 0.05. The analysis was conducted at a confidence level of 95%. As the results of the variance analysis on harvest efficiency and successfully harvested rate carried out for each species gave P-values of 0.0065 and 0.0003, each of which is less than 0.05, the significance level, we can say the kind of species has an effect on the harvest efficiency and the successfully harvested rate. As the results of the variance analysis on harvest efficiency(HE) and pepper left on plant rate(PLR) carried out for helix rotational speed also gave P-values of 0.0038 and 0.0004, respectively, each of which is less than 0.05, the significance level, helix rotation speed can be said to have an effect on harvest efficiency and pepper left on plant rate. The reciprocal action between the species and the helix rotation speed was found to have no effect on harvest efficiency, successfully harvested rate, and pepper left on plant rate.

4. Conclusions

In the present study, two small pepper harvester attached to a walking cultivator (PHAWC) prototypes were designed and produced. The two prototypes were designed and produced to be different in terms of their main frame material, the form of their dividers and picking guides, as well as their helix rotation speeds.

Field performance tests were conducted using two species of chili pepper, the AR Legend and Jeokyoung. These were the species considered most suitable for mechanization and as such were used as the test samples. The tests were repeated three times with various helix rotation speeds; a maximum speed of 500 rpm was used for the first PHAWC prototype, and speeds of 510, 610, and 730 rpm were used for the second PHAWC prototype. The Jeokyoung species of pepper was found to be most suitable for mechanization with both PHAWC prototypes as its harvest efficiency was higher and its pepper left on plant rate and ground fall loss rate were lower.

The results of comparing the first and second PHAWC prototypes are as follows: At the same helix rotation speed of 500 to 510 rpm, for the AR Legend pepper, the average harvest efficiency of the second prototype was higher by 2.2%, and the average pepper left on plant rate and the average damage rate was lower by 2.1% and 3.9%, respectively. In the case of the Jeokyoung pepper, the average harvest efficiency of the second prototype was higher by 5.2%, and the average damage rate was lower by 8.8%, showing that the second prototype has the better performance of the two. However, as the second prototype’s pepper with twig rate was higher and the successfully harvested rate was lower the first prototype, there is still room for improvement, and it is thought that an additional study focusing on solving these issues is warranted. Additional field tests also need to be carried out with more diverse pepper species other than just the species of AR Legend and Jeokyoung.

The second PHAWC shows better performance (higher harvest efficiency and lower damage rate) than the first PHAWC and the product developed in previous literature. Additional performance testing in various field conditions is needed to improve the second PHAWC. This improved chili pepper harvester can be supplied to farms to increase the production of chili pepper.