Firmness measurement of peach by impact force response

Firmness measurement of peach by impact force response first-author corresp ‡Corresponding Author ‡Corresponding Author †E-mail:jwang@zju.edu.cn †E-mail:jwang@zju.edu.cn jwang@zju.edu.cn Department of Biosystems Engineering, Zhejiang University, Hangzhou 310029, China Department of Biosystems Engineering, Zhejiang University, Hangzhou 310029, China The impact force response of a peach impacting on a metal flat-surface was considered as nondestructive determination of firmness. The objectives were to analyze the effect of firmness, drop height, fruit mass, and impact orientation on the impact force parameters, and to establish a relationship between the impact force parameter and firmness. The effect of fruit firmness, drop height and fruit mass on the impact force parameters (coefficient of restitution, percentage of energy absorbed, and coefficient of force-time) was evaluated. The study found that the coefficient of restitution, percentage of energy absorbed, and force-time impact coefficient were significantly affected by fruit ripeness, but not affected by drop height, impact position (fruit cheek), and mass. The percentage of absorbed energy increased with ripeness, while the force-time impact coefficient and coefficient of restitution decreased with ripeness. Relationships were obtained between the three impact characteristic parameters (force-time impact coefficient, coefficient of restitution, and percentage of energy absorbed) and peach firmness using a polynomial model ( R 2 =0.932), S model ( R 2 =0.910), and exponential model ( R 2 =0.941), respectively. Firmness Peach Impact force Coefficient of restitution Percentage of absorbed energy INTRODUCTION 1 Quick and nondestructive methods for measuring texture are critical for controlling postharvest quality of agricultural and food products. Many methods, including force deformation analysis (Schmilovitch et al., 1995 1995 ; Timm et al., 1996 1996 ; Fu et al., 2008 2008 ), vibration (Abbott and Massie, 1995 1995 ; Peleg, 1999 1999 ), acoustics (Sugiyama et al., 1998 1998 ; Wang and Teng, 2006 2006 ), and ultrasound (Mizrach et al., 1997 1997 ; Wang et al., 2006 2006 ), have been applied in determining the quality of fruits and vegetables. Most of these inspection methods are nondestructive and can be conducted with small force applications and displacements. The impact force inspection method, however, has advantages compared to others, because there is no need for placing a transducer on the surface of the product, nor are any cumbersome devices required to immobilize it. Research has shown that impact force analysis (IFA), in which fruit is made to impact a stationary force sensor, can determine firmness on a wide range of crops including peaches, apricots, kiwifruits, blueberries, tomatoes, and pears (Delwiche et al., 1987 1987 ; 1989 1989 ; 1996 1996 ; Scorza et al., 2004 2004 ; Wang et al., 2007 2007 ; Zhang et al., 1994 1994 ). IFA has a number of advantages. First, there is rapid measurement, as each impact lasts only for milliseconds. Second, impact sensors are inexpensive, and finally samples are easy to handle. While prototype online systems have been developed, none have been commercialized. Handling and orienting the fruit at the grader speeds has proven to be a more difficult problem to solve than initially suspected and the technique seems only suitable for quasi-spherical fruits where orientation is not a significant problem (Delwiche et al., 1987 1987 ; 1989 1989 ; 1996 1996 ; Scorza et al., 2004 2004 ; Wang et al., 2007 2007 ; Zhang et al., 1994 1994 ). The IFA measurements are affected by the fruit mass and impact velocity. In principle, measuring fruit mass and dropping the fruit from a fixed height for IFA firmness measurement is simple. On the line grading situation at processing speeds of several fruits per second, attaining a fixed repeatable drop height may not be easy and mass measurement is not simple. There is, therefore, an attendant increase in firmness measurement error, and it becomes necessary to know the acceptable tolerances, or how to correct for changes in these parameters. In addition, the radius of curvature of the fruit also affects IFA measurements, though for regularly shaped quasispherical fruit the radius is closely related to fruit mass, especially if fruit density is constant between fruit (Delwiche et al., 1987 1987 ; 1989 1989 ; 1996 1996 ; Zhang et al., 1994 1994 ). Hence it is very important in this study that the employed evaluation parameters were unaffected by mass and dropping height during IFA measurements. Although firmness information may be improved for greater drop heights (fruit behavior is viscoelastic, which raises the possibility that IFA firmness estimates may change significantly with more energetic impacts and more closely match firmness measured by standard destructive measures like the penetrometer), the fruit can be damaged during higher energy impacts, and it is thus necessary to keep the drop height below the level at which damage is likely. In determining an optimal drop height for IFA measurements, the drop height must be sufficient to provide precision in the force measurements during an impact, but not likely to cause damage. The specific objectives of the research are: (1) analysis of the effect of firmness, drop height, fruit mass, and impact orientation on the impact force parameters, and (2) establishment of a relationship between the impact force parameter and firmness. MATERIALS AND METHODS 1 Peach samples 2 The test samples (Yidianhong peach) were hand harvested in June 3~17, 2007 from an experimental orchard in the Department of Horticulture (30°10′ N, 120°07′ E), Zhejiang University, China. Twenty peaches each day (a total of 300 peaches) were picked on the morning of the test day. Each day, upon arrival at the laboratory, peaches were again inspected to ensure that they were undamaged and not attacked by worm. After inspection, experiments were conducted within 2 h at room temperature (about 18~22 °C). Experimental system 2 The experimental setup consisted of three load cells, signal amplifier, a personal computer (PC) and software to control the experimental setup and to analyze its results. A schematic diagram of the system is presented in Fig. 1 1 . A/D: analog to digital The determination of impact force characteristic required a load cell attached to a mass considerably larger than a peach. A general property of load cells is that they produce a signal not only under compression or tension but from torques as well. When a single load cell is fitted with a platform for bouncing the peach, different points of contact on the platform will cause different torques (Delwiche et al., 1996 1996 ); hence, no reproducible results. To eliminate the torque effect, we used three load cells (Model CA-108, China-YEC Inc., Yongzhou, China; sensitivity 3.24 mV/N, rise time 10 −7 s) spaced triangularly and mounted between two aluminum circular plates (a base plate and a platform plate). The impact force response was detected by an amplifier and a commercial analog to digital (A/D) board connected to the PC, which simultaneously served as the data acquisition system. An optical sensor was used to trigger the acquisition. The signals were sampled at a rate of 20 000 samples/s per sensor for a period of 60 ms. When a spherical object is dropped onto a flat horizontal surface, the force on that surface as a function of time depends on the mass of the sphere, the distance of fall, and the elastic properties of the spherical object. The interrelationships of these variables are not simple, but most are measureable dynamically. Contact force as a function of time f ( t ), however, is acquired through a computer-controlled data acquisition system, and an approach based solely on f ( t ) is highly desirable. Typical f ( t ) curves for multiple bounces in unripe and ripe peaches are shown in Fig. 2 2 . F 1 , F 2 : firmness; f p1 , f p2 : peak force Preliminary tests were performed in order to identify a proper drop height, which did not cause any damage to peach tissue (after 24 h by visual inspection). As a result of preliminary tests, two drop heights of 0.5 cm and 1 cm were selected for all further tests. Drop impacts were conducted at two locations (orientation: cheek 1, cheek 2) along the largest circumference of the fruit. All experiments were conducted using different fruit and each experiment was performed on 10 replicates (i.e., 10 fruits were used per experiment). Firmness measurement 2 After the drop impact experiments, the Magness-Taylor (M-T) firmness of each peach (firmness of destructive penetration) was measured at two orientation points (cheek 1 and cheek 2) in the peach equator by using an Instron 5543 Universal Testing Machine. Penetration tests were conducted with the diameter 8 mm diameter by penetrating in 10 mm (Wang and Teng, 2006 2006 ). The loading rate of the crosshead was 25 mm/min. The M-T firmness was determined to be the mean of the maximum force for two orientation points. Impact force response 2 In this paper, the impact force of a peach is considered with the coefficient of restitution, the percentage of energy absorbed, and the coefficient of impact force-time. Impact peak force, contact time, time-to-peak force, and energy parameters were defined similar to previous researchers, i.e., Wan et al.( 1997 1997 ) and Yen and Wan ( 2003 2003 ), Wang et al.( 2006 2006 ; 2007 2007 ). 1. Typical characteristics of force-time impact curves 3 In Fig. 2 2 , the impact characteristics are shown for two peaches with equal mass but different firmness when dropped from the same height. Fig. 2a 2a shows that the bounce time ( t ) between the two bounces increased with higher firmness. In Fig. 2b 2b , the first impact of the peaches is shown. During the process of the peach contacting the platform, the impact force increased from zero to peak value, then declined quickly. 2. Force-time impact coefficient 3 Fig. 2b 2b shows the peak force f p and the time-to-peak force t p . The force-time impact coefficient c is defined as: c = f p / t p . 3. Coefficient of restitution 3 The coefficient of restitution is equal to the ratio of the impulse (force multiplied by time) of the second impact to the impulse of the first impact. Actually, the coefficient of restitution, r , is equal to v 2 / v 1 , considering two consecutive bounces: r = v 2 / v 1 , where v 1 is the velocity of peach before impact, and v 2 is the velocity of peach after impact. Assuming that the interval between the end of the first impact and the beginning of the second impact was ‘ t ’, and ignoring the resistance of air, when v 2 = g ×( t /2), then we can get $$r = \frac{{v_2 }}{{v_1 }} = \frac{{g(t/2)}}{{\sqrt {2gh} }} = t \times \sqrt {\frac{g}{{8h}}}$$ , where h is the drop height, and g is acceleration due to gravity. 4. Percentage of energy absorbed 3 After the first impact and rebound, the potential energy is equal to the mechanical energy. The amount of energy absorbed E a during impact was calculated from the difference between energy at impact and rebound: E a = mg ( h − h 1 ), where h 1 is the rebound height obtained at time t , and h 1 =[ g ( t /2) 2 ]/2. The percentage of energy absorbed should be the ratio between energy absorbed and initial energy: $$E = \frac{{mg(h - h_1 )}}{{mgh}} \times 100\% = \frac{{(h - h_1 )}}{h} \times 100\%$$ . Statistical analyses 2 The statistical analysis system (SAS) version 6.12 (SAS Cary, NC) was used to determine F values for an analysis of variance (ANOVA) of ripeness, mass, drop height, and impact position (cheek) versus each of the coefficients. RESULTS AND DISCUSSION 1 Effect on peach impact characteristic 2 To evaluate the effect of peach mass [(95.2±2.43), (120.8±3.21), (145.8±4.78) g; three size classes: small, medium, big], firmness [(21.33±3.11), (46.87±3.56), (72.87±4.35) N; three ripe classes: unripe, ripe, overripe] , drop height (0.5, 1 cm), and impact orientation (cheek 1, cheek 2) on the impact force characteristic, statistical analyses of the experimental results were performed using the SAS/STAT procedure (SAS, Institute Inc., USA), and are summarized in Tables 1 1 ~ 3 3 . The magnitudes of the F values indicated the relative effects. Data are expressed as mean± SD (10 fruits repeated). Drop height: h =1 cm; Impacting surface: cheek 1; Impact characteristic parameters are omitted under drop height 0.5 cm and cheek 2 Firmness/ripeness (N) Mass (g) Coefficient of restitution r Percentage of absorbed energy E (%) Force-time impact coefficient f p / t p (N/ms) 18.21~24.10 (21.33±3.11) 95.2±2.43 0.389±0.029 72.90±4.37 6.92±0.43 120.8±3.21 0.358±0.053 74.30±8.31 6.57±0.57 145.8±4.78 0.319±0.048 73.00±8.34 7.02±0.63 42.10~49.18 (46.87±3.56) 95.2±2.43 0.498±0.031 65.80±3.40 10.67±0.92 120.8±3.21 0.479±0.020 65.75±2.65 9.98±0.72 145.8±4.78 0.415±0.047 62.90±4.10 11.31±0.81 66.87~78.97 (72.87±4.35) 95.2±2.43 0.604±0.053 58.31±6.98 16.85±0.98 120.8±3.21 0.581±0.061 60.74±3.29 15.90±1.32 145.8±4.78 0.549±0.062 61.35±5.30 15.80±1.33 Data are expressed as mean± SD (10 fruits repeated). Mass: (120.8±3.21) g; Firmness: (46.87±3.56) N; Impact characteristic parameters are omitted under other ripeness and mass Drop height Impacting surface Coefficient of restitution r Percentage of absorbed energy E (%) Force-time impact coefficient f p / t p (N/ms) 1.0 Cheek 1 0.498±0.041 64.70±4.40 9.87±0.82 1.0 Cheek 2 0.499±0.043 65.55±3.65 9.78±0.92 0.5 Cheek 1 0.455±0.047 63.90±5.20 10.11±0.86 0.5 Cheek 2 0.464±0.053 63.31±5.93 10.15±0.91 Significant at probability P < 0.05 level r : coefficient of restitution; E : percentage of absorbed energy; f p / t p : force-time impact coefficient 2 Element 3 F value r E f p / t p Ripeness 21.17 * * 70.34 * * 284.90 * * Mass 0.11 0.19 2.62 Drop height 1.88 1.03 2.02 Cheek 0.71 2.16 1.68 The mean values are shown in Table 1 1 on the impact characteristic parameters for each mass and degree of ripeness. The percentage of absorbed energy increased, while the force-time impact coefficient and coefficient of restitution decreased with peach ripeness (Table 1 1 ). Drop height and cheek are shown to have little effect on the impact parameters (Table 2 2 ). The peak force divided by the time-to-peak force f p / t p was significantly affected by ripeness. The reduction in peak force with increased maturity was found also by Lichtensteiger et al.( 1988 1988 ). The time to reach the peak force t p was inversely proportional to drop height and mass and increased with ripeness. The peak impact force f p increased, as expected, with drop height and fruit mass, but decreased with ripeness. The f p / t p parameter was not affected by drop height, or impact position (cheek). This is consistent with the finding of Wang et al.( 2006 2006 ) for impact dynamic resonance frequencies of peach. It is possible, however, that other drop heights (e.g., greater drop heights) do affect f p / t p (Teng, 2002 2002 ). The f p / t p parameter was not affected by fruit mass. This agreed with the finding of Brusewitz et al.( 1991 1991 ) for peach. Actually, the peak impact force f p and the time to reach the peak force t p increased with drop height and fruit mass. Being independent of fruit mass makes f p / t p an important impact parameter to indicate fruit firmness. The coefficient of restitution was statistically significantly affected by fruit ripeness. From Eq.( 3 3 ), the coefficient of restitution is related with the ratio of the second impact impulse and the first impact impulse. Actually, energy absorbed by the peach upon impact increased with drop height and fruit mass. Increasing drop height and mass increases the input energy, which leads to more absorbed energy and potentially more internal failure. However, the percentage of absorbed energy was not affected by fruit mass, cheek and drop height. But this does not suggest that other drop height and mass (e.g., greater drop height and mass) do not affect percentage of absorbed energy. The percentage of absorbed energy was significantly affected by the ripeness. The percentage of absorbed energy increased with the ripeness. At the low drop heights, there were greater differences in percentage of absorbed energy between two degrees of ripeness. This indicates that percentage of absorbed energy is also a good indicator of fruit ripeness. Relationship between impact characteristic parameter and firmness 2 Because mass, drop height, and cheek did not significantly affect the three impact characteristic parameters (Table 3 3 ), fruit ripeness (M-T firmness) was only considered in establishing relationship between the impact characteristic parameter and fruit firmness. The relationships between the three impact characteristic parameters and firmness are shown in Fig. 3 3 . The S model, polynomial model, and exponential model were used to fit relationships between the impact characteristic parameters and firmness. Fig. 3 3 shows that the coefficient of force-time and the coefficient of restitution increased with higher firmness and the percentage of absorbed energy decreased with higher firmness. In Table 4 4 the coefficients of determination are given for three different models fitting the ripeness-impact characteristic parameters. It can be seen from the table that the relationship of coefficient of force-time versus firmness was best fitted by the polynomial model, the relationship of coefficient of restitution versus firmness was best fitted by the S model, and the relationship of percentage of absorbed energy versus firmness was better fitted by exponential model. Relationship S model $$y = a + {{(b - a)} \mathord{\left/{\vphantom {{(b - a)} {\left( {1 + {\text{e}}^{\frac{{x - c}}{d}} } \right)}}} \right.\kern-\nulldelimiterspace} {\left( {1 + {\text{e}}^{\frac{{x - c}}{d}} } \right)}}$$ Polynomial model y = a + bx + cx 2 Exponential model $$y = a + b{\text{e}}^{\frac{{ - x}}{c}}$$ Fitted model Force-time impact coefficient ( f p / t p ) a =33.782 b =4.5122 c =83.341 d =14.171 R 2 =0.912 a =9.159 b =−0.280 c =0.0047 R 2 =0.932 a =3.819 b =0.3146 c =−21.074 R 2 =0.881 Polynomial model 5 Coefficient of restitution ( r ) a =0.5107 b =−2.786 c =−21.575 d =15.521 R 2 =0.910 a =0.095 b =0.012 c =0.001 R 2 =0.885 a =−26320.550 b =26320.833 c =−7790677.332 R 2 =0.722 S model 5 Percentage of absorbed energy ( E ) a =55.753 b =600.76 c =−38.704 d =18.405 R 2 =0.938 a =95.70349 b =−1.12117 c =0.00808 R 2 =0.893 a =55.64728 b =61.87476 c =19.06214 R 2 =0.941 Exponential model CONCLUSION 1 Peaches were dropped onto a platform from heights of 0.5 and 1.0 cm and the force versus time curves for the first and second impacts were acquired. Three coefficients were calculated from these curves and the relationships between these coefficients and ripeness, mass, drop height, and impact position (cheek) were determined. The following are specific conclusions of this study: References JA Abbott , DR Massie . Nondestructive dynamic force/deformation measurement of kiwifruit firmness ( Actinidia deliciosa ) . Trans ASAE , 1995 . 38 ( 6 ): 1809 - 1812 . . Nondestructive dynamic force/deformation measurement of kiwifruit firmness (Actinidia deliciosa) GH Brusewitz , TG McCollum , X Zhang . Impact bruise resistance of peaches . Trans ASAE , 1991 . 34 ( 3 ): 962 - 965 . . Impact bruise resistance of peaches MJ Delwiche , T McDonald , SV Bowers . Determination of peach firmness by analysis of impact forces . Trans ASAE , 1987 . 30 ( 1 ): 249 - 254 . . Determination of peach firmness by analysis of impact forces MJ Delwiche , S Tang , JJ Mehlschau . An impact force response fruit firmness sorter . Trans ASAE , 1989 . 32 ( 1 ): 321 - 326 . . An impact force response fruit firmness sorter MJ Delwiche , H Arevalo , J Mehlschau . Second generation impact force response fruit firmness sorter . Trans ASAE , 1996 . 39 ( 3 ): 1025 - 1033 . . Second generation impact force response fruit firmness sorter X Fu , Y Ying , Y Zhou , L Xie , H Xu . Application of NIR spectroscopy for firmness evaluation of peaches . J Zhejiang Univ Sci B , 2008 . 9 ( 7 ): 552 - 557 . 18600785 DOI: 10.1631/jzus.B0720018 http://doi.org/10.1631/jzus.B0720018 . Application of NIR spectroscopy for firmness evaluation of peaches MJ Lichtensteiger , RG Holmes , MY Hamedy , JL Blaisdell . Impact parameters of spherical viscoelastic objects and tomatoes . Trans ASAE , 1988 . 31 ( 2 ): 595 - 602 . . Impact parameters of spherical viscoelastic objects and tomatoes A Mizrach , U Flitsanov , Y Fuchs . An ultrasonic nondestructive method for measuring maturity of mango fruit . Trans ASAE , 1997 . 40 ( 4 ): 1107 - 1111 . . An ultrasonic nondestructive method for measuring maturity of mango fruit K Peleg . Development of a commercial fruit firmness sorter . J Agric Eng Res , 1999 . 72 ( 3 ): 231 - 238 . DOI: 10.1006/jaer.1998.0367 http://doi.org/10.1006/jaer.1998.0367 . Development of a commercial fruit firmness sorter Z Schmilovitch , A Zaltzman , A Hoffman , Y Edan . Firmness sensor and system for date sorting . Appl Eng Agric , 1995 . 11 ( 4 ): 555 - 560 . . Firmness sensor and system for date sorting R Scorza , W Anger , D Peteson , 等 . F Laurens , K Evans . Non-Destructive Evaluation of Post-harvest Peach Fruit Softening . XIth Eucarpia Symposium on Fruit Breeding and Genetics , 2004 . : Angers, France 269 - 273 . ISHS Acta Hort 663 . Non-Destructive Evaluation of Post-harvest Peach Fruit Softening J Sugiyama , T Katsurai , J Hong , H Koyama , K Mikuriya . Melon ripeness monitoring by a portable firmness tester . Trans ASAE , 1998 . 41 ( 1 ): 121 - 127 . . Melon ripeness monitoring by a portable firmness tester B Teng . Studies on the Characteristics of Frequency of Pear and Peach , 2002 . : Hangzhou, China Department of Agricultural Engineering, Zhejiang University , 23 - 35 . PhD Dissertation (in Chinese) . Studies on the Characteristics of Frequency of Pear and Peach EJ Timm , GK Brown , PR Armstrong , RM Beaudy , A Shifazi . Portable instrument for measuring firmness of cherries and berries . Appl Eng Agric , 1996 . 12 ( 1 ): 71 - 77 . . Portable instrument for measuring firmness of cherries and berries YN Wan , MH Yen , C Ay . Nondestructive maturity inspection of fruits and vegetable by impact . J Agric Machinery , 1997 . 6 21 - 33 . . Nondestructive maturity inspection of fruits and vegetable by impact J Wang , B Teng . Firmness evaluation by drop impact characteristics for peach . Int J Food Prop , 2006 . 9 ( 3 ): 439 - 451 . DOI: 10.1080/10942910600596324 http://doi.org/10.1080/10942910600596324 . Firmness evaluation by drop impact characteristics for peach J Wang , B Teng , Y Yu . The firmness detection by excitation dynamic characteristics for peach . Food Control , 2006 . 17 ( 5 ): 353 - 358 . DOI: 10.1016/j.foodcont.2004.12.001 http://doi.org/10.1016/j.foodcont.2004.12.001 . The firmness detection by excitation dynamic characteristics for peach J Wang , TJ Ying , KC Cheng . Evaluation of pear firmness by dynamic characteristics of drop impact . J Sci Food Agric , 2007 . 87 ( 8 ): 1449 - 1454 . DOI: 10.1002/jsfa.2781 http://doi.org/10.1002/jsfa.2781 . Evaluation of pear firmness by dynamic characteristics of drop impact M Yen , Y Wan . Determination of textural indices of guava fruit using discriminate analysis by impact force . Trans ASAE , 2003 . 46 ( 4 ): 1161 - 1166 . . Determination of textural indices of guava fruit using discriminate analysis by impact force X Zhang , ML Stone , D Chen , NO Maness , GH Brusewitz . Peach firmness determination by puncture resistance, drop impact, and sonic impulse . Trans ASAE , 1994 . 37 ( 2 ): 495 - 500 . . Peach firmness determination by puncture resistance, drop impact, and sonic impulse 10.1631/jzus.B0920108 Project supported by the National Natural Science Foundation of China (No. 30570449) and the Program for New Century Excellent Talents in Chinese University (No. NCET-04-0544) 2009-04-18 2009-11-03 2009-10-12 2022-01-05 Copyright © 2009, Journal of Zhejiang University Science 2009 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 12 10