SEPARATING SUNFLOWER SEEDS OUT FROM RECEPTACLES BY TWO AIR-JETS IN STATIC, LINEAR, AND ROTATIONAL MOVEMENT STATES
DOI:10.30825/5.EJPAU.186.2020.23.2

Amir Hossein Mirzabe¹, Gholam Reza Chegini², Jafar Massah², Javad Khazaei^2
1 Department of Mechanical Engineering of Biosystems, College of Agriculture & Natural Resources, University of Tehran, Tehran, Iran
² Department of Mechanical Engineering of Biosystems, College of Aboureihan, University of Tehran, Tehran, Iran

ABSTRACT

In order to eliminate some problems of traditional and mechanical methods of separating sunflower seeds from the receptacles, a new method based on impingement jets was innovated. The effect of the seeds' location, reservoir pressure, nozzle diameter, and the distance between the two nozzles on the separating area of sunflower receptacle, in three different states including, static, linear movement of sunflower receptacle, and rotational movement of nozzles were examined. Also theoretical areas of covered region by the two nozzles in the three different states were calculated. The regions separated by impingement of the air-jet were photographed and the area of the separated regions was calculated based on image processing technique. Also, for the three states, power consumption of compressor was calculated. Moreover, in each state, response surface methodology was used to find the five best optimized points. In the static state, maximum and minimum values of area of separated regions in the central, middle, and side region of sunflower receptacle were equal to 2.63 and 1.39, 3.83 and 1.96, and 4.02 and 2.12 cm², respectively. In the linear movement state, maximum and minimum values of area of separated regions in the central, middle, and side region of sunflower receptacle were equal to 9.78 and 5.72, 16.19 and 10.02, and 17.43 and 11.15 cm², respectively. The corresponding values for rotational movement state were 9.33 and 5.14, 128.98 and 86.12, and 241.89 and 157.65 cm², respectively. Obtained results by variance analyzing showed that in all three states and in each region, effects of independent input factors were significant at probability level of 1%. Also in all tests, no seed was observed to be damaged due to air-jet impinging.

Key words: Sunflower; Image processing; Impingement jet; Power consumption; Optimization.

NOMENCALTURE
			Power	Theoretical power consumption, kW
A	Area of nozzle exit, m2		P-Power	Predicted power consumption by RSM, kW
AI	Angle of impingement, °		R	Real gas constant (R= 8.314 J K-1 mol-1)
ASR	Area of separated region , cm2		R-Power	Real or theoretical power consumption calculated based on the formula, kW
C	percentage of dead space in compressor		RVN	Rotational velocity of nozzles, rpm
CR	Side region		RSM	Response surface methodology
DBNS	Distance between nozzle and sunflower, mm		SR	Side region
DNO	Distance between nozzles to each other, mm		TACR	Theoretical area of covered region by jet, cm2
k	Specific heat transfer (CP / CV)		Vl	Local speed of sound, m sec-1
L	Lubrication factor of compressor		Z	Compressibility factor of air
LVS	Linear velocity of receptacle, cm sec-1		Z1	Compressibility factor of air in input of compressor
?	Mass flow rate of air, kg sec-1		Z2	Compressibility factor of air in output of compressor
MR	Middle region		γ	Specific heat transfer of polytrophic process
ND	Nozzle diameter, mm		ηC	Volumetric flow efficiency of compressor
P	Reservoir pressure, KPa		ηM	Mechanical efficiency of compressor
P1	air pressure in in input of compressor		ηP	Polytrophic efficiency
P2	air pressure in in output of compressor		ρs	Density of supply air, Kg m-3

INTRODUCTION

Sunflower (Helianthus annuus L.) seed is one of the most important oil crops which can also be a valuable source of protein and unsaturated fatty acids in absence of cholesterol [5, 31]. Sunflower is divided into two types, agricultural and ornamental. Agricultural sunflowers are used for two purposes: food consumption and oil extraction. Although food consumption of kernels is one of the major reasons of cultivation of sunflower, the main reason of its cultivation is oil extraction from sunflower seeds [24]. For both purposes, in the process of harvesting sunflowers, the seeds should be separated out from sunflower receptacle (SRL).

Nowadays, the separating seeds out from receptacles is done using traditional manual (in underdeveloped and many developing countries) or mechanized methods (in some developing and developed countries). The efficiency of the traditional manual methods is very low and depends upon the efficiency and experience of the farm workers [8]. In many developing countries, due to traditional farming, low level of farmer’s income, complexity of thresher mechanisms, complex settings of thresher machine and also damage to seeds during operation with thresher machines, farmers are not interested in using mechanized thresher machines [23]. However, to eliminate a number of problems in traditional and mechanical methods of separating sunflower seeds out from their receptacle, a new method should be adopted.

In the present work, as an innovative and new method, using air-jet impingement technology is suggested to separate sunflower seeds from their heads. Today air-jet impingement and water-jet impingement have many applications in industry, food sciences, and agricultural machinery. Air impingement technology is gaining popularity in food-processing operations such as baking, freezing, drying, and toasting [2, 4, 15, 16, 37, 40]. Although there are many applications of jet impingement, but one of its particular and successful applications is separation of the materials from each other [20, 29, 34, 35].

In industry, air- and water-jet are used to de-colorize, ring debug, and clean dirty surfaces. But this type of application of the jets is also exists in agriculture. There is little published literature on application of the air jet impingement on removing arils of pomegranate fruits and extracting of citrus juice and juice sacs [1, 9, 12, 13, 27, 32, 33].

Proper performance of jets in separating the materials from each other in industry and, more importantly, its success in removing arils of pomegranate and extracting the citrus juice and juice sacs, made us think about separating sunflower seeds out from their receptacle by air-jet impingement. First, by one nozzle, the effects of air pressure, nozzle diameter, distance between the nozzle and the surface of the receptacle, jet’s angle of impingement, and sunflower seeds location in static state (nozzle and also receptacle were motionless) were examined [22].

After that, in the second step, effects of air-jet impingement parameters were examined in linear movement. So, we designed and built an apparatus to examine the effects of reservoir pressure, nozzles diameter, distance between nozzle and receptacle surface, and linear velocity, in linear movement state. In this case, one motionless nozzle was used while receptacles had linear movement.

Then, in the third step, the preliminarily model of sunflower separator machine based on air-jet designed and constructed and also, the effects of sunflower seeds location, distance between nozzle and receptacle, angle of impingement, and rotational velocity of receptacle on percentage of separated seeds were examined [24]. In design of the preliminarily machine, one motionless nozzle was used and the receptacles had a rotational movement.

In all previous machines and tests, one nozzle was used. The final separator machine must be able to take a sunflower with all its seeds from one side, and deliver the empty flower and separated seeds on the other side of the machine. So, in design of the final machine, two, three, or even more nozzles are needed. Therefore, even though the effects of nozzle diameter, air pressure, seeds’ location, angle of impingement, distance between nozzle and receptacle surface, linear velocity, and rotational velocity were investigate, due to different formation of turbulent flows, the obtained results of the previous works are not usable and generalizable in design of final machine. After the collision of jet and receptacle surface, some turbulent flows are generated; these turbulent flows go out of the surface, parallel with the receptacle surface. But, in using more than one nozzle, turbulent flows of each nozzle collide to each other, when they are leaving the surface. These collisions have really important effects on the jet performance and efficiency. So, it is necessary to do some experiments, using more than one nozzle, to examine the effects of collision of the adjacent nozzles.

The aim of the present study was to investigate the effects of the seeds’ location, reservoir pressure, nozzle diameter, and distance between nozzles on the area of the separated region when two nozzles were used and 1) the sunflower receptacle and the two nozzles were fixed, 2) the sunflower receptacle had linear movement and the two nozzles were fixed and 3) the sunflower receptacle had rotational movement and the two nozzles were fixed. Also the theoretical area of the covered region by air-jet in the three different states were calculated and compared to experimental results. Moreover, theoretical power consumption of compressor was calculated. Then, in each state, based on theoretical power consumption and obtained results of area of separated region by jets, response surface methodology (RSM) was used and the five best optimized points were determined. Furthermore, predicted powers of compressor by RSM and theoretical ones were compared.

MATERIALS AND METHODS

1. Sampling preparation
Three varieties of sunflowers, namely “Dorsefid”, “Shamshiri” and “Sirena” widely cultivated in Iran, were used in the present study. The “Dorsefid” and “Shamshiri” varieties are native of Iran. The varieties were planted on April 27th, 2012 in research farms of the University of Tehran, located in Pakdasht, Tehran, Iran (longitude of 35.47° N, latitude of 51.67° E, average annual Precipitation 110 mm from 2000 to 2010, height above the sea level of 1025 m, average annual temperature of 18.0 °C from 1993 to 2010). The sunflowers of the three varieties were harvested manually in late September, when receptacles were completely matured.

In order to study the effect of air-jet impingement parameters on area of the separated region in different locations of seed on receptacle, the selected receptacles are divided into three regions, namely central region (CR), middle region (MR) and side region (SR), as shown in Fig. 1. Previous studies conducted on measuring picking force of sunflower seeds from their head [23] and variation of physical and dimensional properties of seeds on different locations on sunflower head [26] led authors to divide each sunflower head into three mentioned regions.

Fig. 1. Three different regions of receptacle, central region (CR), middle region (MR), side region (SR).

2. Static state
A schematic diagram of the experimental setup used to evaluate the effects of operating parameters of two nozzles of air-jet impingement on separating sunflower seeds, in static state is shown in Fig. 2.

Fig. 2. Schematic of experimental set up used to evaluate the effects of air-jet impingement on separating sunflower seeds in static. (1) frame, (2) clamping mechanism, (3) receptacle, (4) two nozzles, (5) vertical guide shaft, (6), horizontal guide shaft (7) changing angle mechanism, (8) valve, (9) pneumatic hoses.

This setup works as follows:

In order to hold the receptacles during tests, a clamping mechanism was used (Fig. 2). The most important part of the clamping mechanism is steel plate with 3 mm in thickness and 360 mm in diameter, so that the mechanism can hold (keep) the receptacles with different diameters, three grooves on the steel plate with 14 cm in length and 5 mm width were created (Fig. 3). The angles between the grooves to each other were 120°. Under each groove and the distance of 30 mm of the underside of the steel plate, a guide rod with 10 mm in diameter, was welded to the plate (Fig. 3). The holder clamp had a hole with 10.5 mm in diameter inside which the guide rod was located. In fact, guide rods and created grooves on the steel plate provide the paths in which the holder clamps can move and hold receptacles with 60 to 360 mm in diameter. To fix the position of holder clamps on the guide rods, firming M6 screws were used so that when the firming screw was loose, holder clamp could be easily moved, but when the screw was tightened, the friction between the guide rod and the screw tip prevents the holder clamps from movement.

A. B.

Fig. 3. Schematics of mechanisms used to change the angle of impingement and maintain the sunflower head during the tests (Mirzabe et al., 2014) A – Schematics of setting angle of impingement mechanism B – Schematics of the sunflower head maintainer mechanism

The distance between the receptacles and the nozzle outlet were set using vertical guide shaft. A horizontal guide shaft was used to examine the effect of air-jet parameters on the area of the separated region in different locations of seed on receptacle.

Although in the present work effect of angle of impingement was not examined, it mechanism works as follow: two semi-circular steel plates with 100 mm in diameter were used. On each steel plate, an imaginary circle with radius of 40 mm was considered and in the center of the imaginary circle a hole with 6 mm diameter was created; also on the circumference of the imaginary circle, 7 holes with 6.2 mm in diameter were created (Fig. 3). The angles between the adjacent holes were 30°. Then the plates were placed at the 35-mm distance from and parallel to each other, so that the created holes on the plates were exactly placed opposite to each other, and two steel plates welded together by a small steel plate. On the both sides of the wind pipe, two small rods with 6 mm in diameter were welded. At the distance of 40 mm from the center of the small rods, two M6 bolts were welded on both sides of the wind pipe. The small rods were located in the created hole on the center of imaginary circle. To avoid angle of impingement changes, two firming M6 screws were used (Fig. 3). Angle changes were avoided by fixing M6 screws to holes (created holes on the circumference of the imaginary circle) and respective two M6 bolts. More detailed information about performance of the machine is presented in [22].

The pressure of the air increases in piston compressor; the high pressure flow of the air is transferred to experimental setup using pneumatic hoses. A switching valve was used to connect and disconnect the air flow. All experiments were performed with three repetitions and the duration of each test, i.e. the duration that the switching valve was open and air jet impinged the seeds, was decided to be 2 seconds.

The harvested flowers of Dorsefid variety were classified into different categories based on value of receptacle diameter and if the value of receptacle diameter ranged from 240 to 260 mm, it was used for the tests.

To conduct tests in the central region (CR) and ensure that the area affected by the jets is only the central region, the tests was conducted in such a way that the central point of the receptacle was placed exactly between the two nozzles. In the middle and side regions, when the distance between the nozzles were equal to 10, 13, 16, 24 mm, the two nozzles were located in one side of the central point of the receptacle; while, when the distance between the nozzles was equal to 112 (in MR) and 200 mm (in SR), the central point of the receptacle was placed exactly between the two nozzles. In 112 and 200 mm cases, the two nozzles were placed in the respective regions (MR or SR); in these cases, the distances between nozzles were far more than the other cases (10, 13, and 16 mm) and output jets from the nozzles cannot be mixed. One-side and two-side location of nozzles are shown in Fig. 4.

A. B.

Fig. 4. One and two-side nozzles to investigate the effect of collision and non-collision of jets, respectively. (A) One-side nozzles, both first and second nozzles are located in one side of the sunflower head center, (B) Two-side nozzles, first nozzle is located in one side of the sunflower head center and second nozzle is located in the other side. A – One-side nozzles B – Two-side nozzles

3. Linear movement state
A schematic diagram of the experimental setup used to evaluate the effects of the operating parameters of two nozzles on separating sunflower seeds out from their receptacles, in linear movement of receptacle is shown in Fig. 5.

Fig. 5. Schematic of experimental set up used to evaluate the effects of air-jet impingement on separating sunflower seeds in linear movement. (1) frame, (2) electromotor, (3)conveyor, (4) receptacle, (5) two nozzles, (6) vertical guide shaft, (7), horizontal guide shaft (8) changing angle mechanism, (9) valve, (10) pneumatic hoses.

This setup works as follows:

In order to make linear movement of nozzles and receptacle relative to each other, rubber conveyer and electromotor were used. The conveyor was made from Viton Rubber tires which are safe for food health. To change the linear velocity of conveyors, the rotational velocity of the electromotor was changed using an invertor (SV 015ic5-1F, LS industrial systems, China). The distance between the receptacles and the nozzles outlet were set using vertical guide shaft. A horizontal guide shaft was used to fix the nozzles location and examine the effect of air-jet parameters on the area of the separated region in different locations of seeds on receptacle. Angle of impingement was set using changing angle mechanism as it was explained, thoroughly before (in the present work effect of angle of impingement was not examined). The pressure of the air increases in piston compressor and the high pressure flow of the air is transferred to experimental setup using pneumatic hoses. A switching valve was used to connect and disconnect the air flow.

The harvested flowers of Shamshiri variety were classified into different categories based on value of receptacle diameter and if the value of receptacle diameter ranged from 270 to 300 mm, it was used for the tests.

In each test, the whole area of the covered region by the air-jet, theoretically, should be in one of the CR, MR, or SR of the receptacle. So, the critical cases were considered as follows: the minimum diameter of receptacle equals to 270 mm and the maximum nozzle diameter equals to 7 mm.

In the central region (CR), the tests were conducted in such a manner that the central point of the receptacle was placed exactly between the two nozzles. In the middle region (MR), the distance between the initial nozzle (the nearest nozzle to the center of receptacle) center and central point of the receptacle was equal to 60 mm; the corresponding value for the side region (SR) was equal to 105 mm. In the critical case, the maximum distance between the nozzles was equal to 15 mm, the maximum distance between the nozzle outlet and receptacle surface was equal to 25 mm, the maximum diameter of the nozzles was equal to 7 mm, the minimum diameter of the receptacle was equal to 270 mm, and the length of linear movement of receptacle was equal to 40 mm. In this case, we can be sure that the whole area theoretically covered by jets is placed in only one region of receptacle (CR, MR or SR).

4. Rotational movement
A schematic diagram of the experimental setup used to evaluate the effects of the operating parameters of two nozzles on separating sunflower seeds, in rotational movement of nozzles state is shown in Figure 6. The main part of the machine consists of the frame, the electromotor, the worm gear box, the clamping mechanism, the mechanism of changing angle of impingement, the mechanism of changing horizontal position of the nozzles, the mechanism of changing distance between nozzles outlet and receptacle surface, the switching valve, the inclined metallic plate, and the transparent polycarbonate talc. This machine works as follows.

Fig. 6. Schematic of experimental set up used to evaluate the effects of air-jet impingement on separating sunflower seeds in rotation. (1) frame, (2) mechanical jack, (3) electromotor, (4) gear box, (5) sloped plate, (6) clamping mechanism, (7), two nozzles, (8) changing angle mechanism, (9) changing horizontal position of nozzles, (10) valve, (11) pneumatic hoses

In order to provide the power required for rotating the receptacle, an electromotor (Hummer, MS632-4, Iran) was used. To regulate the rotational velocity of the electromotor and change the rotational direction, a worm gear box (Taizhou Jiao Xing Transmission Equipment Co., Ltd, RV30, China) was used. In order to hold the receptacles during tests, the receptacles were put on a clamping mechanism. The mechanism was constructed in a way that it could hold the receptacle with the diameter of 6 to 360 mm (before fully explained how it works). Angle of impingement was set using changing angle mechanism as it was explained, thoroughly before (in the current study effect of angle of impingement was not examined). To set the rotational velocity of the metal plate, an invertor (LS industrial systems, SV 015ic5-1F, China) was used.

The harvested flowers of Sirena variety were classified into different categories based on the value of receptacle diameter, and if the value of receptacle diameter ranged from 240 to 260 mm, it was used for the tests.

In the central region (CR), the tests were conducted in such a manner that the central point of the receptacle was placed exactly between the two nozzles. In the middle region (MR), the distance between the initial nozzle (closer nozzle to the center of receptacle) center and central point of the receptacle was equal to 56 mm; the corresponding value for the side region (SR) was equal to 100 mm. In the critical case, the maximum distance between the nozzles was equal to 15 mm, the maximum distance between nozzle outlet and receptacle surface was equal to 25 mm, the maximum diameter of the nozzles was equal to 7 mm and the minimum diameter of the receptacle was equal to 240 mm. In this case, we can be sure that the whole area theoretically covered by jets is placed in only one region of receptacle (CR, MR or SR).

5. Image processing setup
The image processing system, which was used to calculate the area of separated region of receptacle effected by air-jet impingement, consisted of a camera (Canon, IXY 600F, Japan) with 3X IS lens capable of filming up to 120 frames per second (fps) and 12.1 megapixels, four white colored fluorescent lamps (32 W), USB connection, and a laptop computer (DELL, INSPIRON 1558, China) equipped with MATLAB R2012a software package.

The camera was mounted on an image processing box [17]. Each receptacle was put at the center of the camera’s field of view and three metal spheres with the same and identified diameters put at the side of the receptacle. In the first step of processing, one RGB color images was captured from up view of the receptacle and the RGB color space images of calluses were converted into eight-bit gray-scale level. In the second step of processing, the threshold technique was performed to isolate each object from its background. Eight-bit gray-scale intensity represents 256 different shades of gray from black (0) to white (256).

The eight-bit gray-scale images were digitized to binary image using binary transformation on the basis of all the pixels with a brightness level equal to the average of the brightness levels of the three channels [17, 39]. In the third step, the threshold value of the receptacle was determined experimentally [14]. The holes and noise of binary images are filling by morphological closing and opening [14, 17]. From the gray-scale image of Dorsefid variety, pixel values less than 171 were converted to 0 (black), and the values higher than 171 were converted to 256 (white). The threshold levels of the Shamshiri and Sirena varieties gray-scale images were chosen as 169 and 175, respectively. The threshold levels (171, 169, and 175) were determined experimentally [14, 17, 18]. The pixels with value of 256 showed the separated regions of the receptacle, and the pixels with value of 0 showed the remainder [11, 14, 17, 18]. The number of pixels representing the area of separated region of receptacle effected by air-jet impingement was also measured on the captured images using MATLAB R2012a software package. Four main steps in image processing are shown in Fig. 7.

a.   b. c .   d.

Fig. 7. (a) Original RGB color photo of a removed area, (b) eight-bit grayscale photo of a removed area, (c) two-bit binary photo of of a removed area and (d) two-bit binary photo of a removed area with removed noise (Mirzabe and Chegini, 2016)

6. Power consumption
Power consumption in the three static, linear, and rotational states depends on mass flow rate of nozzles. So, at the first step, mass flow rate of nozzles must be known. The critical mass flow rate for a perfect gas with constant specific heat is determined by Eq. (1) [7]:

where: ρs is the density of supply air; A is area of nozzle exit; k is specific heat transfer (CP / CV) and Vl is local speed of sound. In order to calculate supply air density and local speed of sound, the following equations were used [10, 30]:

where: Ts is relative temperature of the fluid; R is real gas constant; k is specific heat transfer; M is molar mass of the air; P is pressure; T is air temperature (thermodynamic temperature); xv is mole fraction of water vapor; Ma is molar mass of dry air (was considered 28.96546 g mol^-1 [30]); Mv is molar mass of water; and  Z  is the compressibility factor.

As you can see in Eq. (1), mass flow rate of the nozzle depends on two changeable parameters, density of supply air and area of the nozzle. Density of supply air depends on air pressure and area of the nozzle depends on nozzle diameter. So, mass flow rate depends on nozzle diameter and reservoir pressure. In the next step, piston compressor efficiency must be calculated, so the following equation was used [6, 38]:

where: ηC is volumetric flow efficiency of compressor; Z1 and Z2 are compressibility factors of air in input and output of compressor; P1 and P2 are air pressure in input and output of compressor; L lubrication factor (L ranges between 0.05 and 0.06 for lubricated compressors and ranges between 0.09 and 0.1 for non-lubricated compressors); C is percentage of dead space (C ranges between 0.06% and 0.12%).

In the last step, power consumption of piston compressor was calculated based on following equation [3, 38]:

where: power consumption is in kW; ηM is mechanical efficiency; Z is compressibility factor; ? is mass flow rate in kg sec^-1; R is real gas constant (R= 8.314 J K-1 mol^-1); γ is specific heat transfer of polytrophic process that ranges from 1.30 to 1.35; P1 and P2 are air pressure in input and output of compressor, and ηP is polytrophic efficiency that can be calculated as:

In order to calculate the supply air density of reservoir and power consumption of compressor, the compressibility factor of the dry air must be known. Compressibility factor of the dry air in different pressures and different temperatures can be calculated using the following equation [21]:

where: P is air pressure; T is air temperature in K; t is temperature in °C. Also, a0, a1 and a2 are constant factors:

Based on the Eq. (1) to (7), power and energy consumption of the compressor were calculated for static, linear, and rotational states. Based on the compressor’s model, constant coefficients were considered to be P1 = 1 bar, Ta = 298.15 K, C = 0.090, L = 0.050, and γ = 1.300, ηM = 0.90, and ηP = 1.2381.

7. Data analysis and design apparatus
In order to choice effective parameters ranges, at first step, scientific sources on air-jet impingements and super-sonic flows were studied. Then, based on the existing formulas and relationships, jet force on sunflower head surface at different air pressure and nozzle diameter levels were calculated theoretically. At second step, some pretests at different levels of effective parameters were done. At the last step, considering mechanical damage to shell of seeds, effective separation of seeds from their head, and low power consumption, choosing factor ranges were done.

In the static state for the Dorsefid variety, the effects of the seeds’ location, reservoir pressure (at 600, 700 and 800 kPa), nozzle diameter (at 5, 6 and 7 mm) and distance between nozzles (at 10, 13, 16 and 24 for CR, at 10, 13, 16 and 112 for MR and at 10, 13, 16 and 200 for SR) on the area of the separated region when the AI and DBNS were equal to 90° and 20 mm, respectively, were studied.

In the linear movement state for the Shamshiri variety, the effects of the seeds’ location, reservoir pressure (at 600, 700 and 800 kPa), nozzle diameter (at 5, 6 and 7 mm) and distance between nozzles (at 10, 13, 16 and 24 for CR, at 10, 13, 16 and 140 for MR and at 10, 13, 16 and 230 for SR) on the area of the separated region when the AI, DBNS, and LVS were equal to 90°, 20 mm, and 2 cm sec^-1, respectively, were studied.

In the rotational movement state for the Sirena variety, the effects of the seeds’ location, reservoir pressure (at 600, 700 and 800 kPa), nozzle diameter (at 5, 6 and 7 mm) and distance between nozzles (at 11, 13, 15 for the three region) on the area of the separated region when the AI, DBNS, and RVN were equal to 90°, 25 mm, and 15 rpm, respectively, were studied.

 The MATLAB R2012a and Microsoft Excel Office 2010 software packages were used for the analysis of the captured photos and the analysis of the data obtained, respectively. Also, the maps of the all parts of machine were first designed in the CATIA V5R20 software package and selection of electromotor, gear box, chains, sprocket wheel, screws, and bearings was done based on engineering calculations and mentioned formulas in mechanical engineering books and handbooks [19, 28, 36].

By using The Design Expert 10 (provided by Stat-Ease Inc) software package analysis of variance carried out to investigate the effects of input factors on ASR. Before and after of the tests, moisture content of the seeds were measured. In all tests, moisture content of the seeds was in range of 26.3 to 31.4% (dry basis). All experiments were performed with three repetitions and the all software packages were run on a DELL 1558 INSPIRON laptop computer.

8. Optimization
Response surface method (RSM) was performed for determining the interaction effects of P and ND parameters. Data were analyzed using the response surface regression procedure and ?t to the second-order quadratic equation. The Design Expert 10 software package was used for data analysis. The ASR and power consumption were considered as response of the test, because the purpose was to design and construct a machine for separating sunflower seeds out from the receptacle with the lowest power consumption. So, in each state, obtained results of ASR and calculated powers consumption of compressor (based on the theory) were used to determine the five best optimized points. For each point, in each region, in each state, optimized ND and P were calculated and power consumption of each case was predicted by software; then, we used these NDs and Ps to calculate theoretical power consumption for optimized points. Finally, predicted (based on the RSM) and theoretical (by formula) power consumption of compressor were compared to each other.

RESULTS

1. Static conditions
1.1. Theoretical area of covered region
In the static conditions the calculated values of theoretical area of covered region by two air-jet impingements for different levels of distances between the nozzles to each other and nozzle diameters are shown in Table 1. Theoretical area of covered region, in three regions of CR, MR, and SR, when DNO were 10, 13, 16, and 24 (or 112 or 200 mm) mm, increased from 2.322 to 3.019, 2.731 to 3.341, 2.864 to 3.775, and 2.864 to 3.775 cm², with increasing ND from 5 to 7 mm, respectively. Also, the calculated values of overlapped area between the two nozzles for different levels of distances between the nozzles to each other and nozzle diameters are shown in Table 1. Results indicated that when DNO was 10 or 13 mm, with increasing ND from 5 to 7 mm, the overlapped area increased from 0.541 to 0.756 and 0.132 to 0.344 cm², respectively.

Table 1. Effect of the seed location, distance between nozzles, nozzle diameter and reservoir pressure on separating sunflower seeds from their receptacle in static state.

Region	DNO (mm)	ND (mm)	Overlapped area (cm2)	TACR (cm2)	Area of separated region (cm2)
					Reservoir pressure (kPa)
					600	700	800
CR	10	5	0.541	2.322	1.39	1.47	1.51
		6	0.649	2.654	1.63	1.71	1.79
		7	0.756	3.019	1.78	1.88	1.97
	13	5	0.132	2.731	1.64	1.74	1.78
		6	0.248	3.055	1.83	1.91	2.02
		7	0.344	3.341	2.01	2.13	2.18
	16	5	0.000	2.864	1.71	1.78	1.87
		6	0.000	3.303	1.95	2.09	2.16
		7	0.000	3.775	2.13	2.22	2.42
	24	5	0.000	2.864	1.83	1.90	1.92
		6	0.000	3.303	2.12	2.24	2.27
		7	0.000	3.775	2.43	2.59	2.63
MR	10	5	0.541	2.322	1.96	2.08	2.20
		6	0.649	2.654	2.22	2.32	2.49
		7	0.756	3.019	2.49	2.67	2.80
	13	5	0.132	2.731	2.34	2.52	2.64
		6	0.248	3.055	2.66	2.81	2.91
		7	0.344	3.341	2.95	3.06	3.21
	16	5	0.000	2.864	2.55	2.69	2.76
		6	0.000	3.303	2.92	3.12	3.21
		7	0.000	3.775	3.36	3.51	3.68
	112	5	0.000	2.864	2.55	2.67	2.81
		6	0.000	3.303	2.98	3.12	3.31
		7	0.000	3.775	3.41	3.61	3.83
SR	10	5	0.541	2.322	2.12	2.31	2.43
		6	0.649	2.654	2.45	2.65	2.81
		7	0.756	3.019	2.76	3.00	3.17
	13	5	0.132	2.731	2.51	2.74	2.89
		6	0.248	3.055	2.82	3.06	3.25
		7	0.344	3.341	3.07	3.34	3.51
	16	5	0.000	2.864	2.65	2.88	3.03
		6	0.000	3.303	3.04	3.32	3.48
		7	0.000	3.775	3.49	3.76	3.98
	200	5	0.000	2.864	2.62	2.91	3.03
		6	0.000	3.303	3.05	3.36	3.51
		7	0.000	3.775	3.50	3.84	4.02

1.2. Experimental results
Results of examining the effect of the operating parameters of two air-jet impingement and seeds location on receptacle on separating sunflower seeds out from the receptacle for Dorsefid variety are shown in Table 1. Results indicated that in three regions with increasing reservoir pressure, nozzle diameter and distance between nozzles, areas of separated regions increased.

Maximum area of separated region in the CR was obtained when the reservoir pressure, nozzle diameter and distance between nozzles were equal to 800 kPa, 7 mm and 24 mm, respectively. Minimum area of separated region was obtained when the pressure, nozzle diameter and distance between nozzles were equal to 600 kPa, 5 mm and 10 mm, respectively. Maximum and minimum values of area of separated regions in the CR were equal to 2.63 and 1.39 cm², respectively (Table 1).

Maximum area of separated region in the MR was obtained when the reservoir pressure, nozzle diameter and distance between nozzles were equal to 800 kPa, 7 mm and 112 mm, respectively.  The corresponding values for SR region were equal to 800 kPa, 7 mm and 200 mm, respectively. Minimum area of separated region in the MR and SR were obtained when the pressure, nozzle diameter and distance between nozzles were equal to 600 kPa, 5 mm and 10 mm, respectively. Maximum and minimum values of area of separated regions in the MR were equals to 3.83 and 1.96 cm², respectively. The corresponding values for SR region were equal to 4.02 and 2.12 cm², respectively (Table 1).

The effects of the distances between the nozzles to each other, nozzle diameters and reservoir pressure on ratio of ASR to TACR (100ASR / TACR) in the three regions of receptacle, in the static conditions are reported in Table 2. The results showed that in all cases, ratio of ASR to TACR increased with increasing reservoir pressure; while, for the other parameters, there is no certain trend.

Maximum value of the ratio of ASR to TACR, in the CR were obtained when the DNO, ND and P were equal to 24 mm, 7 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the CR was obtained when the DNO, ND and P were equal to 16 mm, 7 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the CR were equal to 69.67 and 56.42 %, respectively (Table 2).

Table 2. Effect of the seed location, distance between nozzles, nozzle diameter, and reservoir pressure on separating sunflower seeds from their receptacle in static state

Region	DNO (mm)	ND (mm)	100ASR / TACR (%)
			Reservoir pressure (kPa)
			600	700	800
CR	10	5	59.86	63.31	65.03
		6	61.42	64.43	67.45
		7	58.96	62.27	65.25
	13	5	60.05	63.71	65.18
		6	59.90	62.52	66.12
		7	60.16	63.75	65.25
	16	5	59.71	62.15	65.29
		6	59.04	63.28	65.40
		7	56.42	58.81	64.11
	24	5	63.90	66.34	67.04
		6	64.18	67.82	68.73
		7	64.37	68.61	69.67
MR	10	5	84.41	89.58	94.75
		6	83.65	87.42	93.82
		7	82.48	88.44	92.75
	13	5	85.68	92.27	96.67
		6	87.07	91.98	95.25
		7	88.30	91.59	96.08
	16	5	89.04	93.92	96.37
		6	88.40	94.46	97.18
		7	89.01	92.98	97.48
	112	5	89.04	93.23	98.11
		6	90.22	94.46	100.21
		7	90.33	95.63	101.46
SR	10	5	91.30	99.48	104.65
		6	92.31	99.85	105.88
		7	91.42	99.37	105.00
	13	5	91.91	100.33	105.82
		6	92.31	100.16	106.38
		7	91.89	99.97	105.06
	16	5	92.53	100.56	105.80
		6	92.04	100.51	105.36
		7	92.45	99.60	105.43
	200	5	91.48	101.61	105.80
		6	92.34	101.73	106.27
		7	92.72	101.72	106.49

Maximum value of the ratio of ASR to TACR, in the MR was obtained when the DNO, ND and P were equal to 112 mm, 7 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the MR was obtained when the DNO, ND and P were equal to 10 mm, 7 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the MR were equal to 101.46 and 82.48 %, respectively (Table 2).

Maximum value of the ratio of ASR to TACR, in the SR was obtained when the DNO, ND and P were equal to 200 mm, 7 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the SR were obtained when the DNO, ND and P was equal to 10 mm, 5 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the SR were equal to 106.49 and 91.30 %, respectively (Table 2).

Comparison between experimental results and theoretical calculations showed that for the SR in 19 out of 36 cases and for MR in 2 out of 36 cases, the values of the area of the separated region were more than the area of the theoretical covered region by air-jet; while in the CR in the all cases the values of the area of the separated region were less than the area of the theoretical covered region by air-jet.

Results of analysis of variance conducted in static state are shown in Table 3. Based on obtained results, in the all three CR, MR, and SR, effects of the three independent factors (ND, DNO, and P) were significant at probability level of 1%.

Table 3. Analysis of variance for evaluating the effects of distance between nozzles to each other (DNO), nozzles diameter (ND), and air pressure (p) on area of separated region of sunflower head in static state

Source	DF	CR			MR			SR
		SSE	F-value	P-value	SSE	F-value	P-value	SSE	F-value	P-value
Model	9	3.018	277.04	<0.0001**	7.309	210.99	<0.0001**	7.463	224.79	<0.0001**
DNO (A)	1	1.326	1095.10	<0.0001**	2.872	746.25	<0.0001**	2.164	586.48	<0.0001**
ND (B)	1	1.478	1221.20	<0.0001**	2.977	773.38	<0.0001**	3.137	850.30	<0.0001**
P (C)	1	0.167	137.73	<0.0001**	0.427	110.96	<0.0001**	0.861	233.28	<0.0001**
AB	1	0.063	52.14	<0.0001**	0.092	23.79	<0.0001**	0.051	13.85	0.0010* *
AC	1	0.000	0.10	0.7531	0.005	1.21	0.2812	0.004	1.12	0.3004
BC	1	0.007	5.97	0.0217*	0.006	1.46	0.2376	0.009	2.45	0.1299
A2	1	0.124	102.69	<0.0001**	2.244	583.11	<0.0001**	1.893	513.12	<0.0001**
B2	1	0.004	3.22	0.0842	0.001	0.16	0.6932	0.000	0.01	0.9387
C2	1	0.002	1.41	0.2465	0.000	0.05	0.8215	0.018	4.98	0.0345
Residual	26	0.031			0.100			0.096
Cor. Total	35	3.050			7.409			7.559
Std. Dev.		0.035			0.062			0.061
Mean		1.962			2.845			3.066
C.V. %		1.773			2.181			1.981
R2		0.990			0.986			0.987
Adeq. Precision		65.750			57.231			60.191
* and ** mean significant at levels of 5 and 1%, respectively.

2. Linear movement
2.1. Theoretical area of covered region
In the linear movement of receptacle, the calculated values of theoretical area of covered region by two air-jet impingements for different levels of distance between the nozzles to each other and nozzle diameters are shown in Table 4.

Table 4. Effect of the seed location, distance between nozzles, nozzle diameter, and reservoir pressure on separating sunflower seeds from their receptacle in linear movement state

Region	ND (mm)	DNO (mm)	Overlapped area  (cm2)	TACR (cm2)	Area of separated region (cm2)
					Reservoir pressure (kPa)
					600	700	800
CR	5	10	1.942	11.723	5.72	6.12	6.78
		13	0.333	13.332	6.44	7.12	7.63
		16	0.000	13.665	6.67	7.36	7.81
		24	0.000	13.665	7.01	7.83	8.24
	6	10	2.450	12.455	6.10	6.71	7.11
		13	0.849	14.056	6.84	7.49	8.14
		16	0.000	14.905	7.31	7.84	8.48
		24	0.000	14.905	7.82	8.44	8.92
	7	10	2.957	13.220	6.45	7.04	7.53
		13	1.345	14.831	7.31	7.98	8.33
		16	0.000	16.176	8.01	8.45	9.12
		24	0.000	16.176	8.53	9.27	9.78
MR	5	10	1.942	11.723	10.02	10.55	11.19
		13	0.333	13.332	11.21	12.11	12.90
		16	0.000	13.665	11.50	12.41	13.11
		140	0.000	13.665	12.19	12.91	13.71
	6	10	2.450	12.455	10.72	11.12	12.00
		13	0.849	14.056	11.92	12.72	13.79
		16	0.000	14.905	12.71	13.23	14.65
		140	0.000	14.905	13.37	14.20	15.29
	7	10	2.957	13.220	11.92	12.10	12.97
		13	1.345	14.831	12.78	13.01	14.56
		16	0.000	16.176	14.22	14.57	16.01
		140	0.000	16.176	14.19	15.21	16.19
SR	5	10	1.942	11.723	11.15	11.43	11.77
		13	0.333	13.332	12.93	13.14	13.35
		16	0.000	13.665	13.15	13.47	13.72
		230	0.000	13.665	13.68	14.11	14.46
	6	10	2.450	12.455	12.03	12.27	12.54
		13	0.849	14.056	13.88	14.12	14.32
		16	0.000	14.905	14.71	15.01	15.31
		230	0.000	14.905	15.39	15.61	15.79
	7	10	2.957	13.220	13.02	13.41	13.67
		13	1.345	14.831	14.69	14.88	15.18
		16	0.000	16.176	16.21	16.37	16.81
		230	0.000	16.176	17.03	17.21	17.43

Theoretical area of covered region, in three regions of CR, MR, and SR, when DNO were 10, 13, 16, and 24 (or 140 or 230 mm) mm, increased from 11.723 to 13.220, 13.332 to 14.831, 13.665 to 16.176, and 13.665 to 16.176 cm², with increasing ND from 5 to 7 mm, respectively. Also, calculated values of overlapped area between the two nozzles for different levels of distance between the nozzles to each other and nozzle diameters are shown in Table 4. Results indicated that when DNO was 10 or 13 mm, with increasing ND from 5 to 7mm, overlapped area increased from 1.942 to 2.957 and 0.333 to 1.345 cm², respectively. When the DNO was 16 mm (and more than 16 mm) there was no overlapped area.

2.2. Experimental results
The measured values of area of the separated region of receptacle for the different levels of nozzle diameter, distance between nozzles and reservoir pressure are reported in Table 4. The data showed that in the three region of receptacle, the area of the separated region increased with increasing ND, DNO and P.

Maximum value of ASR in the CR was obtained when the ND, DNO and P were equal to 7 mm, 24 mm and 800 kPa, respectively. Minimum value of ASR in the CR was obtained when the ND, DNO and P were equal to 5 mm, 10 mm and 600 kPa, respectively. Maximum and minimum values of ASR in the CR were equal to 9.78 and 5.72 cm², respectively.

Maximum value of ASR in the MR was obtained when the ND, DNO and P were equal to 7 mm, 140 mm and 800 kPa, respectively. Minimum value of ASR in the MR was obtained when the ND, DNO and P were equal to 5 mm, 10 mm and 600 kPa, respectively. Maximum and minimum values of ASR in the MR were equal to 16.19 and 10.02 cm², respectively.

Maximum value of ASR in the SR was obtained when the ND, DNO and P were equal to 7 mm, 230 mm and 800 kPa, respectively. Minimum value of ASR in the SR was obtained when the ND, DNO and P were equal to 5 mm, 10 mm and 600 kPa, respectively. Maximum and minimum values of ASR in the SR were equal to 17.43 and 11.15 cm², respectively.

In static state of receptacle, the effects of the nozzle diameters, distances between the nozzles and reservoir pressure on ratio of ASR to TACR (100ASR / TACR) in the three regions of receptacle are reported in Table 5. The results showed that in all cases, ratio of ASR to TACR increased with increasing reservoir pressure; while, for the other parameters, there is no certain trend.

Table 5. Effect of the seed location, distance between nozzles, nozzle diameter, and reservoir pressure on separating sunflower seeds from their receptacle in linear movement state

Region	ND (mm)	DNO (mm)	100ASR / TACR (%)
			Reservoir pressure (kPa)
			600	700	800
CR	5	10	48.79	52.21	57.84
		13	48.30	53.41	57.23
		16	48.81	53.86	57.15
		24	51.30	57.30	60.30
	6	10	48.98	53.87	57.09
		13	48.66	53.29	57.91
		16	49.04	52.60	56.89
		24	52.47	56.63	59.85
	7	10	48.79	53.25	56.96
		13	49.29	53.81	56.17
		16	49.52	52.24	56.38
		24	52.73	57.31	60.46
MR	5	10	85.47	89.99	95.45
		13	84.08	90.83	96.76
		16	84.16	90.82	95.94
		140	89.21	94.47	100.33
	6	10	86.07	89.28	96.35
		13	84.80	90.50	98.11
		16	85.27	88.76	98.29
		140	89.70	95.27	102.58
	7	10	90.17	91.53	98.11
		13	86.17	87.72	98.17
		16	87.91	90.07	98.97
		140	87.72	94.03	100.09
SR	5	10	95.11	97.50	100.40
		13	96.98	98.56	100.14
		16	96.23	98.57	100.40
		230	100.11	103.26	105.82
	6	10	96.59	98.51	100.68
		13	98.75	100.46	101.88
		16	98.69	100.70	102.72
		230	103.25	104.73	105.94
	7	10	98.49	101.44	103.40
		13	99.05	100.33	102.35
		16	100.21	101.20	103.92
		230	105.28	106.39	107.75

Maximum value of the ratio of ASR to TACR, in the CR was obtained when the ND, DNO and P were equal to 7 mm, 24 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the CR was obtained when the ND, DNO and P were equal to 5 mm, 13 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the CR were equal to 60.46 and 48.30 %, respectively (Table 5).

Maximum value of the ratio of ASR to TACR, in the MR was obtained when the ND, DNO and P were equal to 6 mm, 140 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the MR was obtained when the ND, DNO and P were equal to 5 mm, 13 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the MR were equal to 102.58 and 84.08 %, respectively (Table 5).

Maximum value of the ratio of ASR to TACR, in the SR was obtained when the ND, DNO and P were equal to 7 mm, 230 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the SR was obtained when the ND, DNO and P were equal to 5 mm, 10 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the SR were equal to 107.75 and 95.11 %, respectively (Table 5).

Comparison between experimental results and theoretical calculations showed that for the SR, in 24 out of 36 cases and for MR, in 3 out of 36 cases, the values of the area of the separated region were more than the area of the theoretical covered region by air-jet (Table 5); while in the CR in the all cases the values of the area of the separated region were less than the area of the theoretical covered region by air-jet (Table 5). Sunflower seeds maturity starts from the side region and gets closer to the center over time. Therefore, the force required to separate the seeds in the SR is less than the MR and CR [23]. Moreover, due to the lower picking force of seeds in the SR and MR compared to the central CR, the turbulent flows that come out parallel to the sunflower surface are able to separate more seeds. Furthermore, the distance between adjacent seeds in SR is more and gets lower by getting closer to the center [26]. Therefore, it is easier for air to penetrate into the space between the seeds and more seeds are separated from the head.

Results of analysis of variance conducted in linear movement state are shown in Table 6. Based on obtained results, in the all three CR, MR, and SR, effects of the three factors were significant at probability level of 1%.

Table 6. Analysis of variance for evaluating the effects of nozzles diameter (ND), distance between nozzles to each other (DNO), and air pressure (p) on area of separated region of sunflower head in linear movement state

Source	DF	CR			MR			SR
		SSE	F-value	P-value	SSE	F-value	P-value	SSE	F-value	P-value
Model	9	31.065	330.45	<0.0001**	78.535	86.44	<0.0001**	92.550	116.22	<0.0001**
ND (A)	1	7.589	726.55	<0.0001**	20.017	198.30	<0.0001**	32.941	372.28	<0.0001**
DNO (B)	1	15.179	1453.22	<0.0001**	34.565	342.41	<0.0001**	48.534	548.51	<0.0001**
P (C)	1	7.483	716.38	<0.0001**	13.256	131.32	<0.0001**	1.309	14.80	0.0007 **
AB	1	0.474	45.33	<0.0001**	0.164	1.63	0.2134	0.959	10.84	0.0029 **
AC	1	0.002	0.15	0.6987	0.025	0.25	0.6243	0.004	0.04	0.8352
BC	1	0.011	1.01	0.3239	0.074	0.74	0.3989	0.000	0.00	0.9498
A2	1	0.000	0.02	0.8820	0.000	0.00	0.9707	0.001	0.02	0.9032
B2	1	1.929	184.66	<0.0001**	20.449	202.57	<0.0001**	29.487	333.25	<0.0001**
C2	1	0.021	1.98	0.1713	0.325	3.22	0.0842	0.000	0.00	0.9499
Residual	26	0.272			2.625			2.301
Cor. Total	35	31.336			81.160			94.851
Std. Dev.		0.102			0.318			0.297
Mean		7.604			12.979			14.257
C.V. %		1.344			2.448			2.086
R2		0.991			0.968			0.976
Adeq. Precision		75.697			38.454			40.635
** means significant at levels of 1%.

3. Rotational movement
3.1. Theoretical area of covered region
In the rotational movement of the two nozzles, the calculated values of theoretical area of covered region by two air-jet impingements for different levels of distance between the nozzles to each other and nozzle diameters are shown in Table 7.

Table 7. Effect of the seed location, distance between nozzles, nozzle diameter, and reservoir pressure on separating sunflower seeds from their receptacle in rotational movement state

Region	ND (mm)	DNO (mm)	Overlapped area  (cm2)	TACR (cm2)	Area of separated region (cm2)
					Reservoir pressure (kPa)
					600	700	800
CR	5	11	1.599	11.120	5.14	5.50	5.83
		13	1.073	13.609	6.03	6.44	6.91
		15	0.296	16.350	6.58	7.02	7.65
	6	11	1.945	11.718	5.87	6.56	6.87
		13	1.481	14.271	6.71	7.14	7.83
		15	0.767	17.075	7.47	7.97	8.47
	7	11	2.290	12.333	6.17	6.66	7.25
		13	1.890	14.949	7.21	7.82	8.84
		15	1.238	17.815	7.88	8.31	9.33
MR	5	11	17.881	102.890	86.12	90.12	92.52
		13	10.318	112.417	95.09	97.75	101.80
		15	2.504	122.195	104.78	106.76	111.99
	6	11	21.745	106.754	90.08	94.86	98.86
		13	14.245	116.344	101.73	105.17	110.54
		15	6.493	126.184	110.24	114.98	119.87
	7	11	25.609	110.618	98.73	101.94	105.50
		13	18.172	120.270	108.04	112.42	116.51
		15	10.483	130.174	121.02	125.05	128.98
SR	5	11	30.674	176.502	157.65	167.73	175.09
		13	17.582	191.558	170.84	183.53	193.50
		15	4.238	206.865	185.93	197.31	209.96
	6	11	37.302	183.131	163.16	177.05	190.30
		13	24.273	198.249	175.50	187.41	208.16
		15	10.993	213.619	188.62	198.49	228.35
	7	11	43.931	189.759	170.53	179.61	204.73
		13	30.964	204.941	181.12	194.19	223.46
		15	17.747	220.374	191.19	204.72	241.89

Theoretical area of covered region, in CR, when DNO were 11, 13, and 15mm, increased from 11.120 to 12.333, 13.609 to 14.949, and 16.350 to 17.815 cm², with increasing ND from 5 to 7 mm, respectively. The corresponding values for MR were 102.890 to 110.618, 112.417 to 120.270, and 122.195 to 130.174 cm², respectively. Also, the corresponding values for SR were 176.502 to 189.759, 191.558 to 204.941, and 206.865 to 220.374 cm², respectively.

Also, calculated values of overlapped area between the two nozzles for different levels of distance between the nozzles to each other and nozzle diameters are shown in Table 7. Results indicated that in CR, when DNO were 11, 13 and 15 mm, with increasing ND from 5 to 7mm, the overlapped area increased from 1.599 to 2.290, 1.073 to 1.890, and 0.296 to 1.238 cm², respectively. The corresponding values for MR were 17.881 to 25.609, 10.318 to 18.172, and 2.504 to 10.483 cm², respectively. Moreover, the corresponding values for SR were 30.674 to 37.302, 17.582 to 30.964, and 4.238 to 17.747 cm², respectively.

3.2. Experimental results
In the rotational movement the measured values of area of the separated region of receptacle for the different levels of nozzle diameter, distance between nozzles and reservoir pressure are reported in Table 8. The data showed that in the three regions of receptacle, the area of the separated region increased with increasing ND, DNO and P.

Table 8. Effect of the seed location, distance between nozzles, nozzle diameter, and reservoir pressure on separating sunflower seeds from their receptacle in rotational movement state

Region	ND (mm)	DNO (mm)	ASR / TACR (%)
			Reservoir pressure (kPa)
			600	700	800
CR	5	11	46.22	49.46	52.43
		13	44.31	47.32	50.78
		15	40.24	42.94	46.79
	6	11	50.09	55.98	58.63
		13	47.02	50.03	54.87
		15	43.75	46.68	49.60
	7	11	50.03	54.00	58.79
		13	48.23	52.31	59.13
		15	44.23	46.65	52.37
MR	5	11	83.70	87.59	89.92
		13	84.59	86.95	90.56
		15	85.75	87.37	91.65
	6	11	84.38	88.86	92.61
		13	87.44	90.40	95.01
		15	87.36	91.12	95.00
	7	11	89.25	92.15	95.37
		13	89.83	93.47	96.87
		15	92.97	96.06	99.08
SR	5	11	89.32	95.03	99.20
		13	89.18	95.81	101.01
		15	89.88	95.38	101.50
	6	11	89.09	96.68	103.91
		13	88.53	94.53	105.00
		15	88.30	92.92	106.90
	7	11	89.87	94.65	107.89
		13	88.38	94.75	109.04
		15	86.76	92.90	109.76

Maximum value of ASR in the CR was obtained when the ND, DNO and P were equal to 7 mm, 15 mm and 800 kPa, respectively. Minimum value of ASR in the CR was obtained when the ND, DNO and P were equal to 5 mm, 11 mm and 600 kPa, respectively. Maximum and minimum values of ASR in the CR were equal to 9.33 and 5.14 cm², respectively.

Maximum value of ASR in the MR was obtained when the ND, DNO and P were equal to 7 mm, 15 mm and 800 kPa, respectively. Minimum value of ASR in the MR was obtained when the ND, DNO and P were equal to 5 mm, 11 mm and 600 kPa, respectively. Maximum and minimum values of ASR in the MR were equal to 128.98 and 86.12 cm², respectively.

Maximum value of ASR in the SR was obtained when the ND, DNO and P were equal to 7 mm, 15 mm and 800 kPa, respectively. Minimum value of ASR in the SR was obtained when the ND, DNO and P were equal to 5 mm, 11 mm and 600 kPa, respectively. Maximum and minimum values of ASR in the SR were equal to 241.89 and 157.65 cm², respectively.

In the rotational movement of the nozzles, effects of the nozzle diameters, distances between the nozzles, and reservoir pressure on ratio of ASR to TACR (100ASR / TACR) in the three regions of receptacle are reported in Table 8. The results showed that in all cases, ratio of ASR to TACR increased with increasing reservoir pressure; while, for the other parameters, there is no certain trend.

Maximum value of the ratio of ASR to TACR, in the CR were obtained when the ND, DNO and P were equal to 7 mm, 13 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the CR were obtained when the ND, DNO and P were equal to 5 mm, 15 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the CR were equal to 59.13 and 40.24 %, respectively (Table 8).

Maximum value of the ratio of ASR to TACR, in the MR was obtained when the ND, DNO and P were equal to 7 mm, 15 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the MR was obtained when the ND, DNO and P were equal to 5 mm, 11 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the MR were equal to 99.08 and 83.70 %, respectively (Table 8).

Maximum value of the ratio of ASR to TACR, in the SR was obtained when the ND, DNO and P were equal to 7 mm, 15 mm and 800 kPa, respectively. Minimum value of the ratio of ASR to TACR, in the SR was obtained when the ND, DNO and P were equal to 6 mm, 15 mm and 600 kPa, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the SR were equal to 109.76 and 88.30 %, respectively (Table 8).

Comparison between experimental results and theoretical calculations showed that for the SR in 8 out of 27 cases, the values of the area of the separated region was more than the area of the theoretical covered region by air-jet (Table 8); while in the MR and CR, in all cases, the values of the area of the separated region was less than the area of the theoretical covered region by air-jet (Table 8).

Results of analysis of variance conducted in rotational movement state are shown in Table 9. Based on obtained results, in the three main regions, effects of the three factors were significant at probability level of 1%.

Table 9. Analysis of variance for evaluating the effects of nozzles diameter (ND), distance between nozzles to each other (DNO), and air pressure (p) on area of separated region of sunflower head in rotational movement state

Source	DF	CR			MR			SR
		SSE	F-value	P-value	SSE	F-value	P-value	SSE	F-value	P-value
Model	9	26.88	157.37	<0.0001**	3156.86	502.97	<0.0001**	10287.22	119.49	<0.0001**
ND (A)	1	8.50	447.85	<0.0001**	957.18	1372.54	<0.0001**	1248.33	130.50	<0.0001**
DNO (B)	1	12.22	643.68	<0.0001**	1900.16	2724.72	<0.0001**	3773.20	394.44	<0.0001**
P (C)	1	5.47	288.01	<0.0001**	278.01	398.65	<0.0001**	4701.27	491.46	<0.0001**
AB	1	0.04	1.91	0.1846	16.59	23.79	<0.0001**	8.00	0.84	0.3731
AC	1	0.19	10.14	0.0054**	0.69	0.99	0.3334	331.91	34.70	<0.0001**
BC	1	0.05	2.47	0.1345	0.68	0.97	0.3383	106.09	11.09	0.0040**
A2	1	0.19	10.05	0.0056**	2.88	4.12	0.0583	0.02	0.00	0.9620
B2	1	0.21	10.82	0.0043**	0.34	0.49	0.4930	0.18	0.02	0.8928
C2	1	0.03	1.40	0.2522	0.34	0.49	0.4930	118.22	12.36	0.0027**
Residual	17	0.32			11.86			162.62
Cor. Total	26	27.21			3168.72			10449.84
Std. Dev.		0.138			0.835			3.093
Mean		7.091			105.609			190.741
C.V. %		1.943			0.791			1.622
R2		0.988			0.996			0.984
Adeq. Precision		49.190			84.596			41.405
** means significant at levels of 1%.

Modeling of ASR using dependent parameters is necessary for determining the interaction effects and predicting values of ASR when values of the dependent parameters change. Quadratic regression applied to model the ASR using dependent parameters includes ND, and DNO and P parameters. Results of linear regression in CR, MR and SR are reported in Table 10. Constant coefficients of following model are presented in Table 10.

Table 10. Constant coefficients for regression models presented to predict area of separated region having values of effective parameters

Coefficients	Static state			Linear movement state			Rotational movement state
	CR	MR	SR	CR	MR	SR	CR	MR	SR
a	2.12	5.86	8.24	8.12	24.50	37.84	7.29	104.83	187.94
b	0.2552	0.4005	0.4141	0.5783	1.04	1.34	0.6872	7.29	8.33
c	0.2619	0.3834	0.3324	0.8863	1.33	1.57	0.8239	10.27	14.48
d	0.0857	0.1517	0.2169	0.5742	0.8481	0.2678	0.5511	3.93	16.16
e	0.0688	0.0734	0.0541	0.1886	0.0977	0.2340	0.0550	1.18	-0.8167
f	0.0212	0.0188	0.0238	-0.0100	0.0394	-0.0156	0.1267	0.2400	5.26
g	-0.0030	0.0166	0.0154	0.0282	0.0657	-0.0045	0.0625	0.2375	2.97
h	-0.0221	0.0087	-0.0017	0.0054	0.0042	0.0129	-0.1783	0.6922	-0.0611
i	-0.1398	-3.10	-5.22	-0.5509	-11.83	-23.80	-0.1850	0.2389	-0.1728
j	-0.0146	-0.0050	-0.0479	-0.0508	0.2017	0.0067	0.0667	0.2389	4.44

4. Theoretical power consumption
Power of two nozzles when the reservoir pressure was equal to 6, 7, and 8 kPa and nozzle diameter was equal 5, 6, and 7 mm was calculated based on Eqs. (1) to (7). Despite the fact that static, linear, and rotational states are really different, but in all three states, we used the same nozzle diameter and reservoir pressure, so the results of power consumption of the three states are the same. These results are shown in Fig. 8. As it can be seen in Fig. 8, an increase in the pressure and nozzle diameter results in an increase in the mass flow rate; an increase in mass flow rate results in an increase in power consumption. Based on the calculations, the minimum and maximum values of power consumption were 0.1173 and 2.8019 kW, respectively.

Fig. 8. Effect of nozzle diameter and reservoir pressure on power consumption of compressor in the three states, static, linear, and rotational

5. Optimized points
Response surface methodology was used to determine optimized points, i.e. certain values for nozzle diameter and reservoir pressure that are able to make a suitable interaction between power consumption and area of separated region. Values of ND, P, predicted ASR, predicted power consumption, and theoretical power consumption, in each region, in static, linear movement, and rotational movement states are shown in Table 11, 12, and 13, respectively.

Table 11. First five optimized points based on response surface methodology for each region in static state

Region	Number of point	P, bar	ND, mm	ASR, cm2	P-Power, kW	Desirability	R-Power, kW	Difference between powers, %
CR	1	6.000	7.000	2.435	1.551	0.711	1.550	0.094
	2	6.008	7.000	2.436	1.556	0.711	1.555	0.084
	3	6.000	6.990	2.432	1.546	0.711	1.545	0.057
	4	6.000	6.982	2.429	1.542	0.710	1.542	0.027
	5	6.150	7.000	2.462	1.652	0.703	1.647	0.275
MR	1	6.000	7.000	3.413	1.551	0.672	1.550	0.094
	2	6.029	7.000	3.418	1.570	0.670	1.568	0.110
	3	6.077	7.000	3.427	1.602	0.666	1.599	0.157
	4	6.000	6.850	3.346	1.481	0.660	1.484	0.192
	5	6.000	6.819	3.332	1.468	0.657	1.470	0.166
SR	1	6.068	7.000	3.529	1.596	0.649	1.594	0.150
	2	6.061	7.000	3.526	1.592	0.649	1.589	0.185
	3	6.081	7.000	3.534	1.605	0.649	1.602	0.181
	4	6.096	7.000	3.540	1.616	0.649	1.612	0.253
	5	6.039	7.000	3.517	1.577	0.649	1.575	0.143

Table 12. First five optimized points based on response surface methodology for each region in linear movement state

Region	Number of point	P, bar	ND, mm	ASR, cm2	P-Power, kW	Desirability	R-Power, kW	Difference between powers, %
CR	1	8.000	5.177	8.697	1.694	0.608	1.687	0.411
	2	8.000	5.186	8.704	1.700	0.608	1.693	0.417
	3	8.000	5.168	8.690	1.688	0.608	1.681	0.404
	4	8.000	5.159	8.682	1.682	0.608	1.675	0.397
	5	8.000	5.202	8.717	1.710	0.608	1.703	0.388
MR	1	7.659	5.377	14.308	1.648	0.576	1.644	0.220
	2	7.664	5.370	14.305	1.646	0.576	1.643	0.207
	3	7.655	5.385	14.312	1.650	0.576	1.647	0.166
	4	7.640	5.400	14.314	1.652	0.576	1.649	0.190
	5	7.650	5.392	14.413	1.651	0.576	1.649	0.120
SR	1	8.000	5.000	17.051	1.576	0.769	1.574	0.148
	2	8.000	5.016	17.054	1.586	0.767	1.584	0.141
	3	7.926	5.000	16.920	1.541	0.763	1.540	0.052
	4	7.895	5.000	16.865	1.526	0.760	1.526	0.015
	5	7.850	5.000	16.785	1.505	0.756	1.506	0.070

Table 13. First five optimized points based on response surface methodology for each region in rotational movement state

Region	Number of point	P, bar	ND, mm	ASR, cm2	P-Power, kW	Desirability	R-Power, kW	Difference between powers, %
CR	1	7.772	5.000	7.799	1.469	0.559	1.471	0.160
	2	7.768	5.000	7.797	1.466	0.559	1.470	0.244
	3	7.758	5.000	7.794	1.462	0.559	1.465	0.217
	4	7.734	5.000	7.785	1.451	0.559	1.455	0.249
	5	7.718	5.000	7.779	1.444	0.559	1.448	0.249
MR	1	8.000	5.000	120.879	1.576	0.662	1.574	0.148
	2	7.987	5.000	120.729	1.569	0.660	1.568	0.080
	3	7.950	5.000	120.321	1.552	0.655	1.551	0.070
	4	8.000	5.086	121.159	1.633	0.655	1.628	0.290
	5	8.000	5.134	121.320	1.665	0.651	1.659	0.352
SR	1	6.173	7.000	215.073	1.668	0.568	1.663	0.318
	2	6.164	7.000	214.946	1.662	0.568	1.657	0.318
	3	6.185	7.000	215.240	1.676	0.568	1.671	0.317
	4	6.195	7.000	215.385	1.683	0.567	1.677	0.335
	5	6.151	7.000	214.764	1.653	0.567	1.648	0.296

In static state, for CR, MR, and SR, values of predicted area of separated region and predicted power consumption for the best point were 2.435 and 1.551, 3.413 and 1.551, and 3.529 cm² and 1.596 kW, respectively. The differences between predicted and calculated power consumption in CR, MR, and SR were less than 0.275, 0.192, and 0.253%. It means that, accuracies of predicted results by RSM are high and the results are valuable and reliable.

Also, in CR, MR, and SR, the maximum and minimum values of desirability were 0.711 and 0.703, 0.672 and 0.657, and 0.649 and 0.649, respectively. In order to compare more readily, results of interaction effects of ND and P on desirability and predicted ASR of the best point (number 1 in Table 11) in each region are illustrated in Fig. 9. For the second, third, fourth, and fifth points, shape of the contour lines were really same as the first point.

A. B. C.

Fig. 9. Interaction effect of nozzle diameter (ND) and reservoir pressure (P) on area of separated region (ASR) and power consumption of compressor (Power) in static state, (A) in central region, (B) in middle region, (C) in side region.

 

In the linear movement state, for CR, MR, and SR, values of predicted area of separated region and predicted power consumption for the best point were 8.697 and 1.694, 14.308 and 1.648, and 17.051 cm² and 1.576 kW, respectively. The differences between predicted and calculated power consumption in CR, MR, and SR were less than 0.417, 0.190, and 0.148%. It means that, accuracies of predicted results by RSM are high and the results are valuable and reliable.

Furthermore, in CR, MR, and SR, the maximum and minimum values of desirability were 0.608 and 0.608, 0.576 and 0.576, and 0.769 and 0.756, respectively. In order to compare more readily, results of interaction effects of ND and P on desirability and predicted ASR of the best point (number 1 in Table 12) in each region are illustrated in Fig. 10. For the second, third, fourth, and fifth points, shape of the contour lines were really the same as the first point.

A. B. C.

Fig. 10. Interaction effect of nozzle diameter (ND) and reservoir pressure (P) on area of separated region (ASR) and power consumption of compressor (Power) in linear movement state, (A) in central region, (B) in middle region, (C) in side region

In the rotational movement state, for CR, MR, and SR, values of predicted area of separated region and predicted power consumption for the best point were 7.799 and 1.469, 120.879 and 1.576, and 215.073 cm² and 1.668 kW, respectively. The differences between predicted and calculated power consumption in CR, MR, and SR were less than 0.249, 0.352, and 0.335%. It means that, accuracies of predicted results by RSM are high and the results are valuable and reliable.

Moreover, in CR, MR, and SR, the maximum and minimum values of desirability were 0.559 and 0.559, 0.662 and 0.651, and 0.568 and 0.567, respectively. In order to compare more readily, results of interaction effects of ND and P on desirability and predicted ASR of the best point (number 1 in Table 13) in each region are illustrated in Fig. 11. For the second, third, fourth, and fifth points, shape of the contour lines were really the same as the first point.

A. B. C.

Fig. 11. Interaction effect of nozzle diameter (ND) and reservoir pressure (P) on area of separated region (ASR) and power consumption of compressor (Power) in rotational movement state, (A) in central region, (B) in middle region, (C) in side region.

DISCUSSIONS

Obtained results in the three states indicated that, with the increase of the nozzle diameter, ASR also increased; theoretically, it can be explained as follows: an increase in the nozzle diameter results in an increase in the cross section of the nozzle outlet, which in turn results in an increase in mass flow rate and therefore an increase in air-jet impingement force in nozzle outlet. This implies an increase in air-jet impingement force on impingement surface [23].

Based on the results in the three states, with the increase of the reservoir pressure, ASR increased; theoretically, it can be explained as follows: an increase in the reservoir pressure results in an increase in the air-jet impingement force in nozzle outlet. This implies an increase in air-jet impingement force on impingement surface and ASR.

Based on the obtained results in the three states, when the two nozzles had overlap, with the increase of the distance between nozzles to each other, ASR also increased; theoretically, it can be explained as follows: an increase in the distance between nozzles results in a decrease overlapped area, so ASR increased. Moreover, obtained results showed that, when the nozzles did not had an overlap, with increasing DNO, ASR increased. It can be explained as: with increasing DNO, collision of two jets, which results in a decrease in the ASR, decreased; so, with increasing distance between nozzles, ASR increased.

The experimental results in the three different states showed that in the same condition (identical pressure, identical nozzle diameter and identical distance between nozzles to each other), the value of the ASR of receptacle on the SR and CR were the most and the lowest, respectively; because for each receptacle, seeds located on the SR of the receptacle reach maturity before the seeds located on the MR of the receptacle. Also, seeds that are located on the MR of the receptacle reach maturity before the seeds located on the CR of the receptacle [23].

Physiological maturity of the receptacles starts from the SR to the CR. So when the receptacle matured, there are immature seeds in CR still absorbing nutrition from the plant; therefore, in most cases, in the CR, maturity does not happen completely and so, picking force of the seeds in CR is more than the MR and SR and value of picking force on MR is more than the SR [25].

Comparison between experimental results and theoretical calculations showed that in the three different states, in some cases, the values of the ASR were more than the TACR by air-jet; It means that in separating sunflower seeds out from the receptacle in the SR and MR, in addition to vertical flows impinged to the receptacle surface, the horizontal flows leave parallel to the surface, which results in separating seeds out from the receptacle. In the CR, picking force of sunflower seeds are more than the picking force in the SR and MR [23]; therefore, the horizontal flows which leave parallel to the surface, cannot separate seeds from the receptacle.

In static state, for Sirena variety, when the DBNS was 20mm, Mirzabe et al. [22] investigated the effect of the angle of impingement in the three levels of 30, 60, and 90°, nozzle diameter in the three levels of 4, 6, and 8 mm, reservoir pressure in the four levels of 5, 6, 7, and 8 bar on ASR. Also, in linear movement state, for Shamshiri variety, we investigated the effects of linear velocity of the receptacle at 1, 2 and 3 cm sec^-1, nozzle diameter at 4, 6 and 8 mm, distance between the nozzle outlet and the receptacle surface at 10, 20 and 30 mm, and reservoir pressure at 600, 700 and 800 kPa on ASR.

A comparison between results of the current work and those works showed that, in the same condition, results of this work were near to the results of those works, while it seemed that because we used two nozzles in the current work, ASR must be doubled; it can be explained as: when an receptacle is under the effect of an air-jet, after collision of jet and receptacle surface, some turbulent flows are generated, these turbulent flows leave the surface, parallel with the receptacle surface. These flows in their pathways separate some seeds; it can be the main reason why the experimental ASR is more than the value of theoretical ASR. When we used two air-jets impingement to separate sunflower seeds from their receptacle, in fact, because of limited dimensions of receptacle, collision of jets is inevitable. Collision of jets causes neutralization of the turbulent flows in the area of receptacle which is located between to nozzles. So collision of jets causes reduction of ASR. So, as the most important result of the current work, it must be mentioned that in design of the final sunflower separator machine based on air-jets impingement, collision of jets must be avoided, as much as possible.

Results of analysis of variance showed that in all three states (static, linear movement, and rotational movement), in each region, three independent input parameters were significant at probability level of 1%. Also, in static state, in all three regions, interaction effect of ND-NDO was significant at probability level of 1%. Moreover, in linear movement, for CR and SR, interaction effect of ND-NDO was significant at probability level of 1%. Furthermore, in rotational movement, interaction effects of ND-P, ND-NDO, and ND-P and DNO-P were significant at probability level of 1% in CR, MR and SR, respectively. Also, in all cases models presented to predict ASR by input parameters were significant at probability level of 1%.

In static state, total power consumption is less than 2.802 kW; while in both linear and rotational movements, total power consumption is more. Despite, extra power needed to make linear or rotational movement of sunflower head, the most of the power consumes by compressor.

CONCLUSIONS

While in all of our previous works, one nozzle was used, in the present study the effects of the two nozzles and distance between nozzles to each other and collision of nozzles were examined.

In the static state for the Dorsefid variety, the effects of the seeds’ location, reservoir pressure, nozzle diameter and distance between nozzles on the area of the separated region were studied. Maximum and minimum values of the ratio of ASR to TACR, in the CR were equal to 69.67 and 56.42 %, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the MR were equal to 101.46 and 82.48 %, respectively; the corresponding values for SR were equal to 106.49 and 91.30 %, respectively.

In the linear movement for the Shamshiri variety, the effects of the seeds’ location, reservoir pressure, nozzle diameter and distance between nozzles to each other on the area of the separated region were studied when the AI, DNS, and LVS were equal to 90°, 20 mm, and 2 cm sec^-1, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the CR were equal to 60.46 and 48.30 %, respectively; the corresponding values for the MR were equal to 102.58 and 84.08 %, respectively; the corresponding values for the SR were equal to 107.75 and 95.11 %, respectively.

In the rotational movement for the Sirena variety, the effects of the seeds’ location, reservoir pressure, nozzle diameter and distance between nozzles on the area of the separated region were studied when the AI, DNS, and RVN were equal to 90°, 25 mm, and 15 rpm, respectively. Maximum and minimum values of the ratio of ASR to TACR, in the CR were equal to 59.13 and 40.24 %, respectively; the corresponding values for the MR were equal to 99.08 and 83.70 %, respectively; the corresponding values for the SR were equal to 109.76 and 88.30 %, respectively.

A comparison between results of this work and previous ones showed that, collision of jets results in neutralization of the turbulent flows in the area of receptacle which is located between to nozzles. So, collision of jets results in reduction of ASR. So, in design of the final sunflower separator machine based on air-jets impingement, collision of jets must be avoided, as much as possible.

Obtained results of theoretical calculation of power consumption showed that in static state, linear movement, and rotational movement compressor consumes less than 2.802 kW. Effective parameters in compressor needed power are nozzle diameter and air pressure. For linear and rotational movements extra power is needed to move sunflower head. Both linear and rotational movement should be considered in design of the final machine. So, power consumed by electromotors to make linear and rotational movement must be summed. So, in design of the final machine, in addition to nozzle diameter and air pressure, power consumption is depended on linear and rotational velocity.

Acknowledgements

The authors would like to thank the University of Tehran for providing technical support for this work. We would also like to sincerely thank Mr. Asghar Mirzabe, Mr. Mahdi Malati, Eng. Ali Javadi and Eng. Javad Yousefi, and deceased Mr. Feizollah Aghasi for their technical help and supervision while writing the paper. We also like to thank Dr. Mohammad Hassan Torabi for his valuable support in editing the language of the paper.

REFERENCES

1. Ando, T., Suzuki, T., Ishii, K., Omura, H., Yamazaki, J., 1988. Apparatus for separating juice sacs of citrus fruits.
2. Banooni, S., Hosseinalipour, S.M., Mujumdar, A.S., Taherkhani, P., Bahiraei, M., 2009. Baking of flat bread in an impingement oven: modeling and optimization. Dry. Technol. 27, 103–112.
3. Bejan, A., 2016. Advanced engineering thermodynamics. John Wiley & Sons.
4. Braud, L.M., Moreira, R.G., Castell-Perez, M.E., 2001. Mathematical modeling of impingement drying of corn tortillas. J. Food Eng. 50, 121–128.
5. Darvishzadeh, R., Pirzad, A., Hatami-Maleki, H., Poormohammad-Kiani, S., Sarrafi, A., 2010. Evaluation of the reaction of sunflower inbred lines and their F1 hybrids to drought conditions using various stress tolerance indices. Spanish J. Agric. Res. 8, 1037–1046.
6. Dixon, S.L., Hall, C., 2013. Fluid mechanics and thermodynamics of turbomachinery. Butterworth-Heinemann.
7. Fay, J.A., Sonwalkar, N., 1996. A fluid mechanics hypercourse. A Fluid Mech. Hypercourse, by James A. Fay Nishikant Sonwalkar. ISBN 0-262-56103-4. Cambridge, Massachusetts, USA MIT Press. May 1996.(Paper).
8. Goel, A.K., Behera, D., Swain, S., Behera, B.K., 2009. Performance evaluation of a low-cost manual sunflower thresher. Indian J. Agric. Res. 43, 37–41.
9. Hayashi, M., Ifuku, Y., Uchiyama, H., Kaga, Y., Nakamori, A., 1981. Apparatus for extracting pulp from citrus fruits.
10. Huber, J.F., 2007. Air jet impingement for levitation.
11. Jodas, D.S., Pereira, A.S., Tavares, J.M.R.S., 2016. A review of computational methods applied for identification and quantification of atherosclerotic plaques in images. Expert Syst. Appl. 46, 1–14.
12. Khazaei, J., Ekrami-Rad, N., Safa, M., Nosrati, S.-Z., 2008a. Effect of air-jet impingement parameters on the extraction of pomegranate arils. Biosyst. Eng. 100, 214–226.
13. Khazaei, J., Massah, J., Mansouri, G.H., 2008b. Effect of some parameters of air-jet on pneumatic extraction of citrus juice and juice sacs. J. Food Eng. 88, 388–398.
14. Koc, A.B., 2007. Determination of watermelon volume using ellipsoid approximation and image processing. Postharvest Biol. Technol. 45, 366–371.
15. Li, A., Walker, C.E., 1996. Cake baking in conventional, impingement and hybrid ovens. J. Food Sci. 61, 188–191.
16. Lujan-Acosta, J., Moreira, R.G., 1997. Reduction of oil in tortilla chips using impingement drying. LWT-Food Sci. Technol. 30, 834–840.
17. Mansouri, A., Fadavi, A., Mortazavian, S.M.M., 2015a. Effects of length and position of hypocotyl explants on Cuminum cyminum L. callogensis by image processing analysis. Plant Cell, Tissue Organ Cult. 121, 657–666.
18. Mansouri, A., Mirzabe, A.H., Ráufi, A., 2015b. Physical properties and mathematical modeling of melon (Cucumis melo L.) seeds and kernels. J. Saudi Soc. Agric. Sci.
19. McCauley, C.J., 2000. Machinery’s Handbook Pocket Companion. Industrial Press Inc.
20. McDermott, W.T., Ockovic, R.C., Wu, J.J., Cooper, D.W., Schwarz, A., Wolfe, H.L., 1994. Surface cleaning using an argon or nitrogen aerosol.
21. Mendes, V.B., Pavlis, E.C., 2004. High-accuracy zenith delay prediction at optical wavelengths. Geophys. Res. Lett. 31.
22. Mirzabe, A.H., Chegini, G.R., 2016. Effect of air-jet impingement parameters on the removing of sunflower seeds from the heads in static conditions. Agric. Eng. Int. CIGR J. 18, 43–59.
23. Mirzabe, Amir Hossein, Chegini, G.R., 2015. Measuring picking force of sunflower seeds and prediction of reasonable range of air-jet parameters to remove sunflower seeds from the head. Agric. Eng. Int. CIGR J. 17.
24. Mirzabe, A.H., Chegini, G.R., Khazaei, J., Massah, J., 2014. Design, construction and evaluation of preliminarily machine for removing sunflower seeds from the head using air-jet impingement. Agric. Eng. Int. CIGR J. 16, 294–302.
25. Mirzabe, A.H., Chegini, G.R., Khazaei, J., Mokhaberi, F., 2013. Determination picking force of sunflower seeds from sunflower head. Adv. Crop Sci. 3, 420–429.
26. Mirzabe, A.H., Khazaei, J., Chegini, G.R., 2012. Physical properties and modeling for sunflower seeds. Agric. Eng. Int. CIGR J. 14, 190–202.
27. Nahir, D., Ronen, B., 1992. Apparatus for removing pulp from fruit.
28. Oberg, E.V., Jones, F.D., Horton, H.L., 1984. Machinery’s handbook: a reference book for the mechanical engineer, desig ner, manufacturing engineer, draftsman, toolmaker and machinist. Industrial Press.
29. Peterson, R. V, Krone-Schmidt, W., 1994. System for precision cleaning by jet spray.
30. Picard, A., Davis, R.S., Gläser, M., Fujii, K., 2008. Revised formula for the density of moist air (CIPM-2007). Metrologia 45, 149.
31. Razi, H., Assad, M.T., 1998. Evaluating variability of important agronomic traits and drought tolerant criteria in sunflower cultivars. JWSS-Isfahan Univ. Technol. 2, 31–44.
32. Sarig, Y., Regev, Y., Grosz, F., 1985. Apparatus for separating pomegranate seeds, scanning apparatus and techniques useful in connection therewith and storage and packaging techniques for separated seeds.
33. Schmilovitch, Z., Sarig, Y., Daskal, A., Weinberg, E., Grosz, F., Ronen, B., Hoffman, A., Egozi, H., 2014. Apparatus and method for extracting pomegranate seeds from pomegranates.
34. Sherman, R., Grob, J., Whitlock, W., 1991. Dry surface cleaning using CO2 snow. J. Vac. Sci. Technol. B Microelectron. Nanom. Struct. Process. Meas. Phenom. 9, 1970–1977.
35. Sherman, R., Hirt, D., Vane, R., 1994. Surface cleaning with the carbon dioxide snow jet. J. Vac. Sci. Technol. A Vacuum, Surfaces, Film. 12, 1876–1881.
36. Shigley, J.E., 2011. Shigley’s mechanical engineering design. Tata McGraw-Hill Education.
37. Sumnu, G., Datta, A.K., Sahin, S., Keskin, S.O., Rakesh, V., 2007. Transport and related properties of breads baked using various heating modes. J. Food Eng. 78, 1382–1387.
38. Ueno, K., Bye, R.E., Hunter, K.S., 2003. Compressor efficiency definitions.
39. Uozumi, N., Yoshino, T., Shiotani, S., Suehara, K.-I., Arai, F., Fukuda, T., Kobayashi, T., 1993. Application of image analysis with neural network for plant somatic embryo culture. J. Ferment. Bioeng. 76, 505–509.
40. XIAO, H., GAO, Z., Lin, H.A.I., YANG, W., 2010. Air impingement drying characteristics and quality of carrot cubes. J. Food Process Eng. 33, 899–918.

Received: 25.04.2020
Reviewed: 3.06.2020
Accepted: 20.06.2020

---

Amir Hossein Mirzabe
Department of Mechanical Engineering of Biosystems, College of Agriculture & Natural Resources, University of Tehran, Tehran, Iran
Telephone: 098 3153239185
Cell phone: 0989399442161
a_h_mirzabe@yahoo.com
email: a_h_mirzabe@alumni.ut.ac.ir

Gholam Reza Chegini
Department of Mechanical Engineering of Biosystems, College of Aboureihan, University of Tehran, Tehran, Iran
Telephone: 098 21 360 406 14
Cell phone: 0989126356329
email: chegini@ut.ac.ir

Jafar Massah
Department of Mechanical Engineering of Biosystems, College of Aboureihan, University of Tehran, Tehran, Iran
Telephone: 098 21 360 406 14
Cell phone: 0989198028454
email: jmassah@ut.ac.ir

Javad Khazaei
Department of Mechanical Engineering of Biosystems, College of Aboureihan, University of Tehran, Tehran, Iran
Telephone: 098 21 360 406 14
Cell phone: 0989123880128
email: jkhazaei@ut.ac.ir

---

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.

---

Main  -  Issues  -  How to Submit  -  From the Publisher  -  Search  -  Subscription