Comparison of balers

The total number of impacts of any intensity (0.1-25 kgF) recorded in the selection and packing line was similar in all four packers (Table 3). This was independent of the duration of the process, since in some balers the process lasts just under 4 min, while in others it lasts around 6 min. The average of the maximum impacts recorded by the sensor was different between the packers (p< 0.05). Packer 1 was the one with the highest value impact. The baler with the maximum impact of lower average was the baler 4. The balers 2 and 3 had intermediate values.

The average of total impacts that the sensor registered in the packers evaluated was different (p< 0.05). The packer 1 had the highest average and the lowest was for packer 4 (Table 3). The low average of the total impacts was due to the fact that the sensor recorded a large number of impacts with values lower than 1 kgF and few impacts greater than 5 kgF.

Comparison between selection and packing line stages

The number of impacts, the maximum and average impact of total impacts recorded by the sensor was different between the packing stages. Stage 2 was where the greatest number of impacts were recorded and in stages 1 and 4 there were fewer impacts (Table 4). Although there were statistical differences (p< 0.05), high intensity impacts can occur at any stage of packaging. However, in stage 1, impacts of higher value occur (Table 4). The mean total impact recorded by the sensor in the evaluated stages was different (p< 0.05). Stage 1 (fruit emptying) was the highest average intensity and stage 2 (rollers and brushes) recorded the lowest average intensity of total recorded impacts. Although in the stage of emptying of fruit has the smallest number of impacts, it is where the fruit suffers more maximum impacts with greater intensity.

Fruit stripping speed

The number of impacts recorded between the speeds evaluated was similar and was between 6.1 impact on the highest speed and 3.5 impacts on the lowest speed. The average maximum impact and the average total impacts were also similar among the three velocities evaluated. The maximum intensity impact was between 6.3 and 8.6 kgF. The average of the total impacts recorded was between 2.4 and 3.9 kgF (p> 0.05). This indicates that, even if the speed of the drainage belt is reduced, the height of falling of the fruits remains the same, so reducing the speed does not reduce the intensity of the impacts.

^{Yull et al. (2015)} reported differences in the number of impacts in three apple packers, with 50, 42 and 28 impacts respectively. In orange packers, they reported 89 impacts on average. Although the packaging process for avocados, apples and oranges has differences, in avocado there were no differences between packers as presented in apple packers. The number of impacts to which both fruits are exposed are similar, around 40 impacts. Although the number of impacts between avocado and apples packers was similar and in oranges was higher, the impact intensity was lower in apples and oranges, as reported by ^{Ferreira et al. (2005)}.

While in avocado the impact of greater intensity was of 9.9 kgf, in apples and oranges barely reached 3 kgf. In another study done in tomato packers, impacts between 5 kgF and 30 kgF were recorded using a triaxial accelerometer (^{Ferreira et al., 2005}). The difference between packers may be due to the sensitivity of the fruits to the impacts due to the anatomical and morphological conformation of the fruit, but also the design of the packaging line.

However, it is not surprising that the packing process in tomato (^{Ferreira et al., 2005}), peach (^{Ahmadi et al., 2010}), apple (^{Yull et al., 2015}) and citrus (^{Miller et al., 1991}) is different. However, it has been reported that the stage of fruit emptying is the stage with high intensity impacts throughout the fruit packing process, similar to what was found in the present investigation. In avocado the stage of emptying of fruits is realized with turning machines that drop the fruits on a cushioned surface of 0.4 m of length and takes to the fruits to a band of rollers of plastic or metal, reason why the shock of the fruits with the rollers causes the impacts. Lowering the speed of the machine reduces the number of impacts, but not the intensity.

Fruits refrigerated for 14 days

At 14 days of cooling the firmness of the fruits was different between treatments at the end of the cooling and at maturity of consumption (p< 0.05). The fruits of treatment 2 presented the lowest firmness at the end of the cooling and at maturity of consumption (mooth metal/0.5 m). The fruits of treatments (0.6 m), coated metal plate (0.6 m), PVC (0.6 m) tubes, fruits (0.5 m) and metal tube (0.5 m) were more firm at the end of the treatment refrigeration and consumption maturity. But treatment 1 was the strongest at maturity of consumption. The treatments covered metal plate (0.4 m), coated metal plate (0.5 m), PVC tubes (0.4 m), fruits (0.6 m), metal tube (0.4 m) and metal tube (0.6 m) in consumption maturity presented a firmness of 1 kgF (Figure 4). Although at the end of the 14 days of refrigeration the loss of mass was similar, at maturity of consumption it was different (p< 0.05), but no treatment exceeded 8% of loss.

Figure 4 Average firmness and mass loss of refrigerated fruits (5 °C ±2 °C and 85% ±10% HR) for 14 days (a); and after being transferred to room temperature (22 °C ±2 °C and 75% ±10% HR) until consumption maturity(b). 

At maturity of consumption, the internal quality of the fruit was similar between treatments (p< 0.05). Metal treatments (0.5 m), coated metal plate (0.4 m), PVC pipes (0.4 m), PVC pipes (0.6 m), fruits (0.6 m) and metal tube (0.4 m) were not damaged in the internal quality. However, the remaining treatments showed darkening and small air cavities in the pulp (Table 5).

Fruits refrigerated for 28 days

At the end of the 28 days of refrigeration and at maturity of consumption the firmness was different (p< 0.05) among the treatments evaluated. At the end of the cooling the firmness of all the treatments was below 4 kgf (close to the firmness of maturity of consumption, 1.5 kgF) and in maturity of consumption the firmness of all the fruits was less of 1.2 kgf, this firmness reached three days after the transfer at room temperature (Figure 5). The loss of mass, at the end of 28 days of refrigeration and at maturity of consumption, was similar between treatments. At the end of the cooling (day 28) the majority of the treatments had exceeded 5% of mass loss and in maturity of consumption exceeded 10%. Although no differences were found in the loss of mass, these percentages at maturity of consumption cause fruit wilt (^{Herrera- González et al., 2013}) (Figure 5). In consumer maturity the internal quality was differentiated between the treatments (p< 0.05). With the exception of treatment 15 (metal tube, 0.6 m), the other treatments presented darkening and air cavities in the fruit pulp in different degrees (Table 5).

Figure 5 Mean of firmness and mass loss of refrigerated fruits (5 °C ±2 °C and 85% ±10% HR) for 28 days (a); and transferred at room temperature (22 °C ±2 °C and 75% ±10% HR) until consumption maturity (b). 

The loss of firmness, mass and internal quality was more evident in the fruits stored for 28 days. The long period in refrigeration and the induced impact affected the physiology of the fruit (^{Pedreschi et al., 2016}), accelerating the enzymatic activity and the degradation of the protopechins (^{Zhou et al., 2007}). Although in pre-harvest, crop management, fertilization and crop maturity confer greater firmness, tolerance to mechanical stress and less dehydration, long periods of refrigeration render the fruit less tolerant and more susceptible to mechanical damage and physiological disorders (^{Pasini et al., 2004}).

Evaluation scale 0= 0%; 0.5= 5%; 1= 10%; 1.5= 15% pulp with darkening and air cavities in the fruit pulp (^{White and Woolf, 2005}).