Computers and Electronics in Agriculture 153 (2018) 54–61 Contents lists available at ScienceDirect Computers and Electronics in Agriculture journal homepage: www.elsevier.com/locate/compag Original papers The inﬂuence of diﬀerent milking settings on the measured teat load caused by a collapsing liner T ⁎ Susanne Dembaa, , Christian Ammona, Sandra Rose-Meierhöferb a b Leibniz Institute for Agricultural Engineering and Bioeconomy, Department of Engineering for Livestock Management, Max-Eyth-Allee 100, 14469 Potsdam, Germany Hochschule Neubrandenburg, University of Applied Sciences, Department of Agricultural Machinery, Brodaer Str. 2, 17033 Neubrandenburg, Germany A R T I C LE I N FO A B S T R A C T Keywords: Machine milking Milking settings Teat-liner interface Pressure Load Milking machine settings such as the machine vacuum, pulsation rate, and pulsation ratio inﬂuence teat tissue and teat condition, but there remains a lack of knowledge about the teat-liner interface and the pressure applied to the teat tissue by the teat cup liner during milking. The aim of the present study was to determine the inﬂuence of diﬀerent milking settings on the teat-liner interface with the help of a pressure-indicating ﬁlm. The Extreme Low Prescale ﬁlm by Fujiﬁlm and a hollow artiﬁcial teat made of silicone were used to measure the inﬂuence of diﬀerent machine vacuum levels (30 kPa, 40 kPa, 50 kPa), diﬀerent pulsation rates (40 cycles min−1, 60 cycles min−1, 80 cycles min−1), and diﬀerent pulsation ratios (60:40, 65:35, 70:30) on the teat load caused by the collapsing liner. The response surface methodology with a central composite design was used to plan the experiment. The experiment was performed with a conventional milking cluster equipped with round silicone liners. The average pressure (AP), the maximum pressure (MP), and the load (L) were used to analyse the inﬂuence of diﬀerent milking settings. Analysis of covariance was used to estimate the diﬀerences between measuring areas, machine vacuum levels, pulsation rates, and pulsation ratios. Machine vacuum levels, pulsation rates, and pulsation ratios had a signiﬁcant inﬂuence on the measured teat load caused by liner collapse; the higher the machine vacuum and the pulsation rate, the higher the measured values of AP, MP, and L. MP values decreased with an increase of the pulsation ratio. The pulsation ratio aﬀected L signiﬁcantly depending on the machine vacuum. The liner applied more pressure to the end of the teat compared with the whole teat barrel. In conclusion, the results of the present investigation show that diﬀerent adjustments to the machine vacuum, the pulsation rate, and the pulsation ratio can signiﬁcantly inﬂuence the pressure applied to the whole teat by a collapsing liner. 1. Introduction Machine milking signiﬁcantly aﬀects teat tissue parameters (Gleeson et al., 2002) and can worsen teat condition (Mein et al., 2001; Mir et al., 2015). Several methods, such as teat scoring (Neijenhuis et al., 2000; Mein et al., 2001; Rose-Meierhöfer et al., 2014), ultrasound measurements (Neijenhuis et al., 2001; Gleeson et al., 2004; Parilova et al., 2011), calculating the Touch Point (TP), the Over-Pressure (OP), and the Liner Compression (LC) (Mein and Reinemann, 2009), and direct pressure measurements (Davis et al., 2001; Tol et al., 2010; Leonardi et al., 2015), are available to assess the inﬂuence of machine milking on teat tissue and teat condition directly after milking as well as over a longer period of time. The settings of machine vacuum, pulsation rate, and pulsation ratio aﬀect teat tissue and teat condition. Excessive machine vacuum leads to ⁎ cracks in the epithelium of the teat tissue (Williams and Mein, 1985). Hamann and Mein (1988) investigated the thickness of the teat end in response to diﬀerent vacuum settings (30 kPa, 50 kPa, and 70 kPa) and found that the teat-end thickness increased as the vacuum level increased; tissue stiﬀness increased as well (Hamann and Mein, 1988). The comparison between a machine vacuum at 30 kPa, 40 kPa, and 50 kPa showed signiﬁcant diﬀerences in teat thickness (Hamann et al., 1993), and a comparison of two diﬀerent vacuum settings showed that milking at a lower level resulted in less colour changes of the teat and less corniﬁcation of the teat oriﬁce (Ebendorﬀ and Ziesack, 1991). According to Ryšánek et al. (2001), a high vacuum correlated signiﬁcantly (correlation coeﬃcient of 0.50) with the formation of teatend hyperkeratosis, and reducing the machine vacuum decreased the risk of hyperkeratosis (Neijenhuis et al., 2005). Reinemann et al. (2001) did not ﬁnd a signiﬁcant correlation between machine vacuum level Corresponding author. E-mail address: sdemba@atb-potsdam.de (S. Demba). https://doi.org/10.1016/j.compag.2018.08.011 Received 12 January 2017; Received in revised form 12 March 2018; Accepted 4 August 2018 0168-1699/ © 2018 Elsevier B.V. All rights reserved. Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. and teat-end callosity, but they found a tendency towards more teats with worse scores and fewer teats with improving conditions with a vacuum of 50 kPa compared to 42 kPa. Parilova et al. (2011) tested the inﬂuence of two diﬀerent vacuum levels (39 kPa and 45 kPa) on the traits teat length, teat diameter at the base, teat diameter at the middle, teat canal length, teat end width, teat wall thickness, and teat cistern width. The authors found a longer teat and teat canal, a narrower teat diameter at the base and at the middle, a wider teat and teat cistern, and a thicker teat wall with a higher machine vacuum. A machine vacuum level of 42 kPa resulted in increased teat wall thickness and a decrease in teat cistern diameter compared with a machine vacuum of 50 kPa (Besier and Bruckmaier, 2016). In contrast, a low machine vacuum level extended the milking duration and worsened the teat end condition (Reid and Johnson, 2003). Diﬀerent pulsation settings can inﬂuence both teat condition and teat tissue condition. Grindal (1988) found that extending the suction phase led to an increase in teat lesions and subcutaneous bleedings. Hansen et al. (2006) investigated the inﬂuence of diﬀerent pulsation rates and pulsation ratios on teat thickness and found signiﬁcant differences between ‘fast’ (22–55 cycles min−1, 66–81% suction phase) and ‘slow’ (47 cycles min−1, 43% suction phase) treatments; the ‘fast’ treatment resulted in an increase in teat thickness. A comparison of seven d-phase duration levels (50 ms, 100 ms, 150 ms, 175 ms, 225 ms, 250 ms, and 300 ms) resulted in a signiﬁcant reduction in the estimated cross-sectional area of the teat canal at d-phase durations of 50 and 100 ms (Upton et al., 2016). Bluemel et al. (2016) found that an extended c-phase during the pulsation cycle decreased the total vacuum per cycle by 1 kPa and increased the opening and closing duration of the liner, so the authors concluded that an extended c-phase indicated gentler milking. In contrast, Gleeson et al. (2004) observed no negative eﬀect on teat tissue by widening the pulsation ratio, and Ferneborg and Svennersten-Sjaunja (2015) detected no negative eﬀects of diﬀerent pulsation ratios on teat-end hyperkeratosis or teat tissue thickness as well. A quarter individual milking system for conventional milking parlours (MultiLactor®) with a machine vacuum of 37 kPa, a sequential pulsation at a rate of 60 min−1 and a pulsation ratio of 65:35 resulted in better teat colour scores after milking compared with a conventional milking system with a machine vacuum of 40 kPa, a pulsation rate of 60 min−1, and a pulsation ratio of 60:40 (Rose-Meierhöfer et al., 2014). Machine vacuum and pulsation settings inﬂuence TP, OP, and LC values as well. TP was found to increase with a higher machine vacuum (Spencer et al., 2007), and according to Mein et al. (2003), OP increased with increasing liner vacuum. These authors found a slight increase in OP as the pulsation c-phase was shortened as well, so adjusting the pulsation settings to a ratio of 65:35 and a rate of 60 cycles min−1 might reduce the eﬀects of OP. A higher claw vacuum was found to create a larger diﬀerence in pressure across the wall of the collapsed liner and result in a higher LC (Mein and Reinemann, 2009). Both OP and LC were found to increase as the vacuum level of the individual liners increased (Reinemann and Mein, 2011). The aim of the present investigation was to determine the inﬂuence of diﬀerent milking settings on the pressure applied to the whole teat by a collapsing liner using a pressure-indicating ﬁlm and a hollow artiﬁcial teat made of silicone. Table 1 Coded and uncoded levels for the independent variables used in the response surface methodology. Independent variable Coded level Milking system vacuum (kPa) Pulsation rate (cycles min−1) Pulsation ratio −1 0 +1 30 40 60:40 40 60 65:35 50 80 70:30 Table 2 The 15 combinations of machine vacuum, pulsation rate, and pulsation ratio detected with the central composite design. Machine vacuum (kPa) Pulsation rate (cycles/min−1) Pulsation ratio 30 30 30 30 30 40 40 40 40 40 50 50 50 50 50 40 40 60 80 80 40 60 60 60 80 40 40 60 80 80 60:40 70:30 65:35 60:40 70:30 65:35 60:40 65:35 70:30 65:35 60:40 70:30 65:35 60:40 70:30 diﬀerent levels of each variable were coded to conform with the CCD (Table 1). The CCD design resulted in 15 unique combinations (Table 2) of machine vacuum, pulsation rate, and pulsation ratio. The central point (40 kPa, 60 cycles min−1, 65:35) was repeated ten times, and ﬁve replicates were performed for each of the other combinations for a total of 80 measurements. 2.2. Data collection Data were collected using an experimental setup similar to Demba et al. (2016). The Extreme Low ﬁlm type (Prescale by Fujiﬁlm; KAGER Industrieprodukte GmbH, Dietzenbach, Germany) and an artiﬁcial teat made of silicone were used to investigate the inﬂuence of diﬀerent milking settings on the teat load caused by a collapsing liner. The pressure range of the ﬁlm was 0.05–0.2 MPa. The teat had a length and a mean diameter of 56 mm and 21 mm, respectively and was hollow with a teat wall thickness of 4.5 mm. According to the manufacturer, the silicone rubber had a Shore A hardness of 25, a density of 1.16 g cm−3 at a temperature of 23 °C, a tensile strength of 5.00 N mm−2, an ultimate elongation of 350%, a tear resistance of more than 20 N mm−1, and a linear shrinkage of 0.5%. The experiment was carried out in the experimental milking parlour of the Leibniz Institute for Agricultural Engineering and Bioeconomy e.V. (ATB). A conventional milking cluster (Surge, GEA Group AG, Düsseldorf, Germany) equipped with round silicone liners (IQPro, GEA Group AG, Düsseldorf, Germany) was used, and each liner had a shaft diameter of 24 mm, a mouthpiece diameter of 21 mm, and a head diameter of 58 mm. All measurements were performed using the same teat cup; the other teat cups were closed with plugs (Fig. 1). The pressure-indicating ﬁlm was cut into pieces (35 mm × 45 mm), all of which were attached with tape to the same position on the teat. The artiﬁcial teat was then inserted in the teat cup so that the collapsed liner and the sides of the pressureindicating ﬁlm were pressed together, and milking was simulated for 1 min. The pieces of ﬁlm were then analysed with FDP-8010E software by Fujiﬁlm (Prescale by Fujiﬁlm; KAGER Industrieprodukte GmbH, 2. Materials and methods 2.1. Study design Following Bade et al. (2009), the response surface methodology (RSM) with the central composite design (CCD) was used to design the experiment. Machine vacuum, pulsation rate, and pulsation ratio were chosen as independent variables, and the three levels of each variable were as follows: the machine vacuum was adjusted at 30 kPa, 40 kPa, and 50 kPa; the pulsation rates were 40 min−1, 60 min−1, and 80 min−1; and the pulsation ratios were 60:40, 65:35, and 70:30. The 55 Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. (AV)i * x is the regression coeﬃcient for the interaction between the measuring area and machine vacuum; and εik is the residual. The inﬂuence of diﬀerent milking settings on MP was calculated using the following model: yik = μ + Ai + V∗x + V2∗x 2 + PL∗z + PR∗w + (VPL) ∗x∗z + εik (2) where yik is the observed values of the i-th measuring area (i = WHOLE, END), and the k-th measurement (k = 1, …, 80) for MP; μ is the overall mean; Ai is the ﬁxed eﬀect of the measuring area (i = WHOLE, END); V is the regression factor of the machine vacuum x; V2 is the regression factor of the squared machine vacuum x2; PL is the regression factor of the pulsation rate z; PR is the regression factor for the pulsation ratio w; (VPL)i * x * z is the regression factor of the interaction between the machine vacuum and the pulsation rate; and εik is the residual. The following model was used to calculate the inﬂuence of the diﬀerent milking settings on L: yik = μ + Ai + V∗x + V2∗x 2 + PL∗z + PR∗w + (AV)i ∗w + (VPR) ∗x∗w + εik where yik is the observed values of the i-th measuring area (i = WHOLE, END), and the k-th measurement (k = 1, …, 80) for L; μ is the overall mean; Ai is the ﬁxed eﬀect of the measuring area (i = WHOLE, END); V is the regression factor of the machine vacuum x; PL is the regression factor of the pulsation rate z; PR is the regression factor of the pulsation ratio w; (AV)i * w is the regression factor of the interaction between the measuring area and the machine vacuum; (VPR) * x * w is the regression factor of the interaction between the machine vacuum and the pulsation ratio; V2 is the regression factor of the squared machine vacuum x2; and εik is the residual. All tests were performed at a signiﬁcance level of 0.05 and all estimated values are given with the standard error of the mean. As a measure of accuracy for the predictions the root mean square error (RMSE) is given. Fig. 1. The experimental setup to measure the teat load caused by liner collapse. Dietzenbach, Germany). The average and maximum pressure on the area of the teat where colour is generated (AP and MP in MPa, respectively) and the load, which is the product of the pressurised surface area and the average pressure (L in N), were used to analyse the inﬂuence of the diﬀerent milking settings. AP, MP, and L were calculated for the whole area covered by the ﬁlm (WHOLE), as well as the area of the teat end (END), which was deﬁned as the area of the lower third of the barrel of the artiﬁcial teat, so WHOLE included END. 3. Results Fig. 2 shows pressure-indicating ﬁlm scans of the tested levels of machine vacuum, pulsation rate, and pulsation ratio. The results of the ANCOVA showed a signiﬁcant inﬂuence of the machine vacuum (P < 0.0001), the squared values of the machine vacuum (P < 0.0275), the pulsation rate (P < 0.0012), and the interaction of measuring area and machine vacuum (P = 0.0008) on AP (Fig. 3). The pulsation ratio did not aﬀect AP (P = 0.111) and therefore was omitted from the regression equations for AP below. The mean AP values diﬀered signiﬁcantly between both measuring areas (P < 0.0001); AP was higher for END compared with WHOLE. The regression equation for AP and WHOLE derived from the results for the eﬀect estimates of the ANCOVA model for AP (R2 = 0.730) is the following: 2.3. Statistical analysis Data were analysed using the SAS 9.4 software package (SAS Institute Inc., Cary, NC, USA). Analysis of covariance (ANCOVA) was used to estimate the diﬀerences between measuring areas and the slopes of the inﬂuence of machine vacuum levels, pulsation rates, and pulsation ratios on the response variables using the MIXED procedure. The null hypotheses for AP, MP, and L were that the regression coefﬁcients were zero, and it was assumed that there were no diﬀerences between the measuring areas of the tested traits. First, the inﬂuence of all factors and all twofold interactions were tested. Then all interactions without signiﬁcant inﬂuences were deleted from the model and the new model was calculated again. The models were compared and the Akaike information criterion (AIC) was used to choose the optimal model. The following model was used to calculate the inﬂuence of the diﬀerent milking settings on AP: yik = μ + Ai + V∗x + V2∗x 2 + PL∗z + PR∗w + (AV)i ∗x + εik (3) V −40 V −40 2 PL−60 ⎞ + 0.0031 ⎛ ⎞−0.0034 ⎛ ⎞ AP = 0.09537 + 0.0124 ⎛ ⎝ 20 ⎠ ⎝ 10 ⎠ ⎝ 10 ⎠ (4) where AP is the average pressure in MPa; V is the machine vacuum; and PL is the pulsation rate. The regression equation for AP and END derived from the results for the eﬀect estimates of the ANCOVA model for AP (R2 = 0.730, RMSE = 0.009) is as follows: (1) V −40 V −40 2 PL−60 ⎞ + 0.0031 ⎛ ⎞−0.0034 ⎛ ⎞ AP = 0.11062 + 0.0188 ⎛ ⎝ 20 ⎠ ⎝ 10 ⎠ ⎝ 10 ⎠ where yik is the observed values of the i-th measuring area (i = WHOLE, END), and the k-th measurement (k = 1, …, 80) for AP; μ is the overall mean; Ai is the ﬁxed eﬀect of the measuring area (i = WHOLE, END); V is the regression factor of the machine vacuum x; V2 is the regression factor of the squared machine vacuum x2; PL is the regression factor of the pulsation rate z; PR is the regression factor of the pulsation ratio w; (5) where AP is the average pressure in MPa; V is the machine vacuum; and PL is the pulsation rate. According to the results of the ANCOVA, the machine vacuum 56 Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. Fig. 2. Pressure-indicating ﬁlm scans, with the teat end at the bottom, of the tested levels of machine vacuum, pulsation rate, and pulsation ratio. (P < 0.0001), the squared values of the machine vacuum (P < 0.0001), pulsation rate (P = 0.0058), pulsation ratio (P = 0.01), and the interaction of machine vacuum and pulsation rate (P = 0.0077) inﬂuenced the MP values signiﬁcantly (Fig. 4). The mean MP values did not diﬀer between WHOLE and END, and no signiﬁcant inﬂuence of the interaction between machine vacuum and area was found. The regression equation for MP and WHOLE derived from the results for the eﬀect estimates of the ANCOVA model for MP (R2 = 0.360, RMSE = 0.01) is the following: the pulsation ratio. The regression equation for MP and END derived from the results for the eﬀect estimates of the ANCOVA model for MP (R2 = 0.360, RMSE = 0.01) is as follows: V −40 V −40 2 PL−60 ⎞ + 0.0029⎛ ⎞−0.00697⎛ ⎞ MP = 0.2031 + 0.0071⎛ ⎝ 20 ⎠ ⎝ 10 ⎠ ⎝ 10 ⎠ SPR−65 ⎞−0.00313VPL −0.0027⎛ (6) 5 ⎝ ⎠ where MP is the maximum pressure in MPa; V is the machine vacuum; PL is the pulsation rate; and SPR is the amount of the suction phase of the pulsation ratio. The results of the ANCOVA showed that the machine vacuum (P < 0.0001), the squared values of the machine vacuum (P < 0.0001), and pulsation rate (P = 0.0009) as well as the interactions of machine vacuum and measuring area (P < 0.0001) and V −40 V −40 2 PL−60 ⎞ + 0.0029⎛ ⎞−0.00697⎛ ⎞ MP = 0.20222 + 0.0071⎛ ⎝ 20 ⎠ ⎝ 10 ⎠ ⎝ 10 ⎠ SPR−65 ⎞−0.00313VPL −0.0027⎛ (7) 5 ⎝ ⎠ where MP is the maximum pressure in MPa; V is the machine vacuum; PL is the pulsation rate; and SPR is the amount of the suction phase of 57 Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. Fig. 3. The measured values of the average pressure (AP in MPa) depending on the machine vacuum level in kPa and the pulsation rate in cycles min−1 for the measuring area of the whole teat (left) as well as the teat end area (right) with pulsation ratios of (a) 60:40, (b) 65:35, and (c) 70:30. Diﬀerent grey intensities show intervals of 0.005 MPa. V −40 V −40 2 PL−60 ⎞ + 4.77 ⎛ ⎞−7.7033 ⎛ ⎞ L = 44.4125 + 9.96 ⎛ 10 ⎝ 20 ⎠ ⎠ ⎝ 10 ⎠ ⎝ PR−65 ⎞−3.1375VPR −2.37 ⎛ ⎝ 5 ⎠ machine vacuum and pulsation ratio (P = 0.0477) inﬂuenced L signiﬁcantly (Fig. 5). The L values were signiﬁcantly higher (P < 0.0001) for WHOLE compared with END. The regression equation for L and WHOLE derived from the results for the eﬀect estimates of the ANCOVA model for L (R2 = 0.865, RMSE = 14.058) is the following: V −40 V −40 2 PL−60 ⎞ + 4.77 ⎛ ⎞−7.7033 ⎛ ⎞ L = 105.8 + 24.82 ⎛ ⎝ 20 ⎠ ⎝ 10 ⎠ ⎝ 10 ⎠ PR−65 ⎞−3.1375VPR −2.37 ⎛ ⎝ 5 ⎠ (9) where L is the load in N; V is the machine vacuum; PL is the pulsation rate; and PR is the amount of the suction phase of the pulsation ratio. 4. Discussion The machine vacuum signiﬁcantly inﬂuenced AP, MP, and L; the values of all three traits increased as the machine vacuum increased. This agrees with Tol et al. (2010), who found an increasing pressure diﬀerence over the liner wall during the d-phase of pulsation with increasing vacuum level. The results of the present investigation led to the assumption that a higher machine vacuum resulted in a higher teat load during milking, and the results of Hamann and Mein (1988) and Hamann et al. (1993) conﬁrmed this assumption. The authors found a (8) where L is the load in N; V is the machine vacuum; PL is the pulsation rate; and PR is the amount of the suction phase of the pulsation ratio. The regression equation for L and END derived from the results for the eﬀect estimates of the ANCOVA model for L (R2 = 0.865, RMSE = 14.058) is as follows: 58 Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. Fig. 4. The measured values of the maximum pressure (MP in MPa) depending on the machine vacuum level in kPa and the pulsation rate in cycles min−1 for the measuring area of the whole teat (left) as well as the teat end area (right) with pulsation ratios of (a) 60:40, (b) 65:35, and (c) 70:30. Diﬀerent grey intensities show intervals of 0.005 MPa. and L, so the values of the three traits were lowest at a machine vacuum of 30 kPa. It should be noted that a certain pressure of the teat cup liner is necessary to massage the teat during the c and d pulsation phases, so both Reid and Johnson (2003) and Besier and Bruckmaier (2016) advised a machine vacuum of no less than 30 kPa. In the present study, the pulsation rate inﬂuenced the values of AP, MP, and L; the three traits increased with an increase in pulsation rate. A higher pulsation rate resulted in a higher teat load due to the teat cup liner, and this could explain the results of Hansen et al. (2006), who found a better teat tissue condition and thus a less stressed teat with a lower pulsation rate. The pulsation ratio signiﬁcantly inﬂuenced the values of MP; the values decreased with an extension of 5 percentage points to the suction phase. Hansen et al. (2006) found signiﬁcant diﬀerences between a large and a small pulsation ratio as well, and short c and d pulsation phases resulted in increased teat thickness and higher stress on the teat. negative inﬂuence of the vacuum level on the condition of the teat tissue in both studies; i.e., the condition of the teat tissue worsened as the machine vacuum level increased. Parilova et al. (2011) found that teat tissue condition worsened with increasing machine vacuum as well. Besier and Bruckmaier (2016) compared three milking vacuum treatments (treatment 1: 42 kPa system vacuum, 33 kPa min claw vacuum during milk ﬂow; treatment 2: 50 kPa system vacuum, 34 kPa claw vacuum during milk ﬂow; treatment 3: 42 kPa system vacuum, during milking claw vacuum drop down to 24 kPa) and found increased teat wall thickness and decreased teat cistern diameter with treatment 2. Teat condition, which indicates the stress on the teat tissue during milking, is inﬂuenced by the machine vacuum. The colour changes of the teat after milking were lower at a lower machine vacuum level (Ebendorﬀ and Ziesack, 1991), and a reduced machine vacuum resulted in less teat-end hyperkeratosis (Ryšánek et al., 2001; Neijenhuis et al., 2005). The lower the machine vacuum, the lower the values of AP, MP, 59 Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. Fig. 5. The measured values of the load (L in N) depending on the machine vacuum level in kPa and the pulsation rate in cycles min−1 for the measuring area of the whole teat (left) as well as the teat end area (right) with pulsation ratios of (a) 60:40, (b) 65:35, and (c) 70:30. Diﬀerent grey intensities show intervals of 20 N. The calculated values of TP, OP, and LC, which are used to estimate the pressure caused by a collapsing liner, were aﬀected by the adjustment of the milking settings as well; TP, OP, and LC increased with increasing machine vacuum (Mein et al., 2003; Spencer et al., 2007; Reinemann and Mein, 2011). Mein et al. (2003) suggested a pulsation rate of 60 cycles min−1 to reduce the eﬀects of the calculated OP values, which could help to reduce the formation of machine-related teatend hyperkeratosis and is consistent with the results of the present investigation. In terms of the pulsation ratio, Mein et al. (2003) conﬁrmed the results of the present study, because they found a slight increase in the calculated OP values with a shortened c pulsation phase. However, the calculated values of TP, OP, and LC were much lower compared with the measured pressure values of the present study. The average OP ranged from < 5 to > 20 kPa (Mein et al., 2003; Mein and Reinemann, 2009) and 9.8–18.2 kPa (Leonardi et al., 2015) depending on liner type. Spencer et al. (2007) found TP values between 25.73 and 33.52 kPa The comparison of diﬀerent d-phase durations indicated a signiﬁcant reduction in the estimated cross-sectional area of the teat canal with shorter durations (Upton et al., 2016). The results of the present study regarding MP disagree with the results of Gleeson et al. (2004), who did not ﬁnd a negative inﬂuence of diﬀerent pulsation ratios on teat load (tissue and teat-end hyperkeratosis). However, the results of the present research could not yet be transferred directly to the live animal because a teat of a dairy cow is much more ﬂexible and reacts diﬀerently to the teat load due to liner collapse than an artiﬁcial teat. Tol et al. (2010) found 2.5 times smaller pressure values on a live cow teat compared with the pressure values of an artiﬁcial teat. The AP values were higher for END compared with WHOLE, which conﬁrms the results of Tol et al. (2010) and Muthukumarappan et al. (1994). The authors of both studies found that the teat cup liner applied more pressure to the teat end than to the whole teat. 60 Computers and Electronics in Agriculture 153 (2018) 54–61 S. Demba et al. depending on the operating hours of a liner. While the estimation of TP, OP, and LC is based on the calculation of pressure diﬀerences across the liner walls, the pressure values of the present research were directly measured and resulted from a force applied to an area. This could be an explanation for the big diﬀerences between the pressure values. According to Reinemann and Mein (2011), LC is the most biologically relevant way to measure the pressure applied to the teat by a liner, but it is not the same as direct pressure measurements with a pressure sensitive ﬁlm and an artiﬁcial teat. The pressure values measured in the present investigation are much higher than those found in earlier studies. Muthukumarappan et al. (1994) measured a pressure of 18–35 kPa, Davis et al. (2001) found pressure values between 20 kPa and 41 kPa, Tol et al. (2010) measured pressure values of 99–180 kPa at the teat end, and Leonardi et al. (2015) detected pressures of 20–34 kPa. The used artiﬁcial teats could be the reason for these ﬁndings because they diﬀer in their dimensions and the liner could have applied a diﬀerent amount of pressure to the teat. The artiﬁcial teat used by Tol et al. (2010) was 20 mm longer, had a tapered shape, and a 2.5 mm-thinner teat wall compared to the artiﬁcial teat used in the present investigation. Diﬀerences in the material of the artiﬁcial teats oﬀer an additional explanation. Muthukumarappan et al. (1994) used a liquid-ﬁlled, ﬂexible, not extensible artiﬁcial teat made of a plastic teat cup plug, a surgical glove ﬁnger, and a cloth glove ﬁnger. The artiﬁcial teat of Davis et al. (2001) was made of natural gum rubber or a gel-like material. The artiﬁcial teat of Leonardi et al. (2015) was covered by a silicone rubber with a Shore A hardness of 10. In the present investigation, an artiﬁcial teat made of silicone rubber with a Shore A hardness of 25 was used. The functionality of the pressure-indicating ﬁlm could be another reason for the higher pressure values of the present investigation. The pressureindicating ﬁlm used in the present investigation does not support shear stress (Rodríguez-Martínez et al., 2012), and shear stress can alter the colour intensity measured by the ﬁlm (Patterson et al., 1997). On the other side neither negative pressure nor bending the sensor around the artiﬁcial teat inﬂuenced the measuring results (Demba et al., 2016). The studies of Henak et al. (2014) and Mutlu et al. (2014) showed that the Prescale pressure-indicating ﬁlm measures a higher maximum pressure compared with other pressure-measurement methods. In conclusion, the results of the present investigation showed that diﬀerent adjustments of the machine vacuum, the pulsation rate, and the pulsation ratio result in diﬀerent teat loads caused by a collapsing liner as measured with Prescale pressure-indicating ﬁlm. The higher the machine vacuum and the pulsation rate, the higher the measured values of AP, MP, and L. MP values decreased with an increase of the pulsation ratio. The pulsation ratio aﬀected L signiﬁcantly depending on the machine vacuum. Proceedings of the ASAE Annual International Meeting 2001, Sacramento, California. Demba, S., Elsholz, S., Ammon, C., Rose-Meierhoefer, S., 2016. The usability of a pressureindicating ﬁlm to measure the teat load caused by a collapsing liner. Sensors 16 (10), 1597. Ebendorﬀ, W., Ziesack, J., 1991. Studies into reduction of milking vacuum (45-kPa) and its impact on teat stress, udder health as well as on parameters of milk-yield and milking. Monatshefte Fur Veterinarmedizin 46 (24), 827–831. Ferneborg, S., Svennersten-Sjaunja, K., 2015. The eﬀect of pulsation ratio on teat condition, milk somatic cell count and productivity in dairy cows in automatic milking. J. Dairy Res. 82 (4), 453–459. Gleeson, D.E., O'Callaghan, E.J., Rath, M.V., 2002. Eﬀect of milking on bovine teat tissue as measured by ultrasonography. Irish Veter. J. 55 (12), 628–632. Gleeson, D.E., O'Callaghan, E.J., Rath, M.V., 2004. Eﬀect of liner design, pulsator setting, and vacuum level on bovine teat tissue changes and milking characteristics as measured by ultrasonography. Irish Veter. J. 57, 289–296. Grindal, R.J., 1988. The role of the milking machine in mastitis. 2. Br. Vet. J. 144 (6), 524–533. Hamann, J., Mein, G.A., 1988. Responses of the bovine teat to machine milking – measurement of changes in thickness of the teat apex. J. Dairy Res. 55 (3), 331–338. Hamann, J., Mein, G.A., Wetzel, S., 1993. Teat tissue-reactions to milking – eﬀects of vacuum level. J. Dairy Sci. 76 (4), 1040–1046. Hansen, S., Hamann, J., Lacy-Hulbert, J., Woolford, M.W., 2006. Long-term eﬀects of diﬀerent pulsation characteristics on teat thickness, teat skin moisture and teat skin pH of dairy cows. Milchwissenschaft-Milk Sci. Int. 61 (3), 239–241. Henak, C.R., Kapron, A.L., Anderson, A.E., Ellis, B.J., Maas, S.A., Weiss, J.A., 2014. Specimenspeciﬁc predictions of contact stress under physiological loading in the human hip: validation and sensitivity studies. Biomech. Model. Mechanobiol. 13, 387–400. Leonardi, S., Penry, J.F., Tangorra, F.M., Thompson, P.D., Reinemann, D.J., 2015. Methods of estimating liner compression. J. Dairy Sci. 98 (10), 6905–6912. Mein, G.A., Neijenhuis, F., Morgan, W.F., Reinemann, D.J., Hillerton, J.E., Baines, J.R., Ohnstad, I., Rasmussen, M.D., Timms, L., Britt, J.S., Farnsworth, R., Cook, N., Hemling, T., 2001. Evaluation of bovine teat condition in commercial dairy herds: 1. Non-infectious factors. In: Proceedings of the 2nd International Symposium on Mastitis and Milk Quality 2001, Lelystad, The Netherlands, pp. 347–351. Mein, G.A., Reinemann, D.J., 2009. Biomechanics of milking: teat-liner interaction. In: Proceedings of the ASABE Conference, American Society of Agricultural and Biological Engineers (Eds.), Reno, Nevada. Mein, G.A., Williams, D.M.D., Reinemann, D.J., 2003. Eﬀects of milking on teat-end hyperkeratosis: 1. Mechanical forces applied by the teatcup liner and responses of the teat. In: National Mastitis Council Regional Meetings Proceedings, Forth Worth, Texas, pp. 114–123. Mir, A.Q., Bansal, B.K., Gupta, D.K., 2015. Short term changes in teats following machine milking with respect to quarter health status in cows. J. Anim. Res. 5 (3), 467. Muthukumarappan, K., Reinemann, D.J., Mein, G.A., 1994. Compressive load applied to the bovine teat by the teatcup liner. In: Proceedings of the ASAE International Winter Meeting, Atlanta, Georgia. Mutlu, I., Ugur, L., Celik, T., Buluc, L., Muezzinoglu, U.S., Kisioglu, Y., 2014. Evaluation of contact characteristics of a patient-speciﬁc artiﬁcial dysplastic hip joint. Acta Bioeng. Biomech. 16, 111–120. Neijenhuis, F., Barkema, H.W., Hogeveen, H., Noordhuizen, J.P.T.M., 2000. Classiﬁcation and longitudinal examination of callused teat ends in dairy cows. J. Dairy Sci. 83 (12), 2795–2804. Neijenhuis, F., Klungel, G.H., Hogeveen, H., 2001. Recovery of cow teats after milking as determined by ultrasonographic scanning. J. Dairy Sci. 84 (12), 2599–2606. Neijenhuis, F., Klungel, G.H., Hogeveen, H., Noordhuizen, J.P.T.M., 2005. Machine milking risk factors for teat end callosity in dairy cows on herd level. In: Hogeveen, H. (Ed.), Mastitis in dairy production: current knowledge and future solutions. Wageningen Academic Publishers, Wageningen, Netherlands. Parilova, M., Stadnik, L., Jezkova, A., Stole, L., 2011. Eﬀect of milking vacuum level and overmilking on cows' teat characteristics. Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis 59 (5), 193–202. Patterson, R., Pogue, D., Viegas, S., 1997. The eﬀects of time and light exposure on contact and pressure measurements using Fuji prescale ﬁlm. Iowa Orthop. J. 17, 64–69. Reid, D.A., Johnson, A.P., 2003. Trouble shooting herds with poor teat condition. In: Proceedings of the NNC Annual Meeting Proceedings, pp. 124–127. Reinemann, D.J., Davis, M.A., Costa, D., Rodriguez, A.C., 2001. Eﬀects of milking vacuum on milking performance and teat consition. In: Proceedings of the AABP-NMC International Symposium on Mastitis and Milk Quality, Vancouver. Reinemann, D.J., Mein, G.A., 2011. Unraveling the mysteries of liner compression. In: Penry, J.F., Mein, G.A., Mansell, P. (Eds.), Proceedings of the Countdown Meeting. Dairy Australia, Melbourne, Australia. Rodríguez-Martínez, R., Urriolagoitia-Sosa, G., Torres-San Miguel, C.R., Hernández-Gómez, L.H., Urriolagoitia-Calderón, G., Carbajal-Romero, M.F., 2012. Development of an experimental apparatus for testing a total knee prostheses focused on mexican phenotype. Int. J. Phys. Sci. 7, 5779–5786. Rose-Meierhöfer, S., Müller, A.B., Mittmann, L., Demba, S., Entorf, A.C., Hoﬀmann, G., Ammon, C., Rudovsky, H.J., Brunsch, R., 2014. Eﬀects of quarter individual and conventional milking systems on teat condition. Prev. Veter. Med. 113 (4), 556–564. Ryšánek, D., Olejník, P., Babák, V., 2001. Vacuum ﬂuctuation in short milk tube during peak milk ﬂow. In: Conference on Physiological and Technical Aspects of Machine Milking, ICAR Technical Series No. 7, Nitra, Slovak Republic, pp. 125–130. Spencer, S.B., Shin, J.W., Rogers, G.W., Cooper, J.B., 2007. Short communication: Eﬀect of vacuum and ratio on the performance of a monoblock silicone milking liner. J. Dairy Sci. 90 (4), 1725–1728. Tol, P.P.J.vd, Schrader, W., Aernouts, B., 2010. Pressure distribution at the teat-liner and teatcalf interfaces. J. Dairy Sci. 93 (1), 45–52. Upton, J., Penry, J.F., Rasmussen, M.D., Thompson, P.D., Reinemann, D.J., 2016. Eﬀect of pulsation rest phase duration on teat end congestion. J. Dairy Sci. 99 (5), 3958–3965. Williams, D.M., Mein, G.A., 1985. The role of machine milking in the invasion of mastitis organisms and implications for maintaining low infection-rates. Kieler Milchwirtschaftliche Forschungsberichte 37 (4), 415–425. 5. Conﬂict of interest The authors declare that they have no conﬂicts of interest. 6. Funding This research did not receive any speciﬁc grant from funding agencies in the public, commercial, or not-for-proﬁt sectors. References Bade, R.D., Reinemann, D.J., Zucali, M., Ruegg, P.L., Thompson, P.D., 2009. Interaction of vacuum, b-phase duration, and liner compression on milk ﬂow rates in dairy cows. J. Dairy Sci. 92 (3), 913–921. Besier, J., Bruckmaier, R.M., 2016. Vacuum levels and milk-ﬂow-dependent vacuum drops aﬀect machine milking performance and teat condition in dairy cows. J. Dairy Sci. 99 (4), 3096–3102. Bluemel, F.E., Savary, P.E., Schick, M.E., 2016. Eﬀects of an extended c-phase on vacuum conditions in the milking cluster. Biosyst. Eng. 148, 68–75. Davis, M.A., Reinemann, D.J., Mein, G.A., 2001. Development and testing of a device to measure compressive teat load applied to a bovine teat by the closed teatcup liner. In: 61

1/--страниц