Experimental kinetic friction coefficient (k) determination in the interface polycarbonate blade and flat rubber-belt when interacting with Ore, Lubricant and Pressure
Jairo A Martins
Department of Bucking Machines Handling and Conveyor Components
Engineeering, Metso Minerals, Av. Independncia, 2500 Iporanga, ZIP
It is well known that even the most advanced Finite Element Analysis (FEA) version is not able to predict friction coefficients among materials. This paper presents an alternative and experimental method based on Design of Experiments (DOE) to determine the friction coefficient between polycarbonate blades against rubber-belt. Those experiments correlate Ore, Lubricating and Pressure in order to determine the friction coefficients in the interface between these materials with polycarbonate and rubber belt. The experiments results are of extreme importance for polycarbonate application on belt conveyor scraper and other applications involving polymeric surfaces under friction.
Keywords:
Friction Coefficient; Polycarbonate; Ore; Rubber Belt; Design of Experiments (DOE)
Reference:
J A Martins; Istvn Kövesdy and J.D.Bressan (2010). “Experimental kinetic friction coefficient (uk) determination in the interface polycarbonate plate and flat rubber-belt when interacting with Ore, Lubricant and Pressure, Int. J. Mining and Minerals Engineering, Vol 2, No 2, pp. 159-168.
(Published with the kind permission of the Int. J. Mining and Minerals Engineering)
1.Introduction
Polymers are becoming more widely employed as bearing and slider materials due to the availability and design flexibility of these materials. As polymers improve its properties and as designers learn the advantages of polymers application, public acceptance also increases. Friction is a very common phenomenon in daily life and in industry, which is governed by the chemical and physical phenomenon occurring in a thin surface layer of solid bodies in the moving contact (Zhang, S.W 1998). With the application of polymers for tribological purposes, increasing and extending to even more new areas, increasing research has been carried out on the Tribology of polymers and its application in various industries. A widespread interest in plastics has grown since the middle 20th century due to the features of their structure, design flexibility, specific mechanical behavior, and considerable possibility to change the polymer properties (Myshkin N.K. et al., 2005). Most studies of contact mechanics have focused on small loads where the contact depends linearly on the load (Bush A W, 1975 and Campana C. and Müser, 2007). Many polymeric materials have an excellent strength-weight ratio and often are easy to manufacture; some polymers also possess excellent tribological properties (Yang C., et. al,2006 and Shooter K., Thomas R.H., 1952). Polycarbonate is a widely used thermoplastic polymer for its high hardness and toughness either as a single phase material or as the matrix of a composite material (Grenwood J. A. and Willianson J. B. P., 1996). There have been fewer theoretical studies on the friction and wear behavior of polymers in general, and polycarbonate in particular (Hyun S. Pei. L, 2004). Glassy polymers are usually different from metals in many aspects, such as a larger elastic zone, elastic modulus dependent on temperature, pressure and strain rate sensibility, strain-hardening due to orientation of polymer chain (orientation hardening), strain softening and a more viscous behavior (Lee, J.H., Xu, G.H., et al., 2001). Due to the difficulty to test and determine friction coefficient in the field/application and neither by Finite Element Analysis these experiments were performed. This paper aims to replicate the scrapers concept used in conveyors but under a laboratory environmental and toward mathematical tool, Design of Experiments (DOE) (Minitab 15- Manual, 2009). This tool was used to plan, analyze and afterwards to determine the friction coefficient as well as the significance of each parameter [iron ore, lubrication (water) and pressure] on its value, occurred in the interface between polycarbonate and rubber-belt. This knowledge is of deep relevance on the design and operation of scrapers; polycarbonate made, and belt conveyors designs.
2.Materials and Method
2.1. Rubber-belt roughness
The measurement of roughness by the Ra parameter was done in order to verify the possible rubber-belt wearing variation before and after the experiments. The roughness measurements were performed on the rubber-belt surface in 9 points in the longitudinal as well as in the transversal direction.
2.2. Special device
In order to carry out the experiments, a special device was developed, according to attached Figure 1 below.
The Load 1 is added on the upper plate to press the polycarbonate blade against the rubber belt and the load 2 is hanging to make the rubber-belt to start the movement. The table where the rubber-belt is fixed is supported by ball bearings and pulled forward by a steel cable which turns the pulley (bearing) and further on engaged in the hanged Load 1. The rubber-belt travel is limited on 200mm. The test speed depends on the Load 2 but the weight value was selected for sliding speeds within the range of 0,01-1,0 cm/s, in which, according to Myshkin N.K, et al. 2005, the friction coefficient is speed-independent.
Figure 1: Device to carry out the friction experiment
2.3. Experimental parameters and DOE
The chosen parameters for the experiments were: Cubic Ore Iron size ranging from 150m to 300m, the Pressure applied by the superior weight (Load 1) was 0,04MPa and 0,10MPa and the Lubricant (water). The study of these variable combinations was done by running a design of experiment DOE method in the Minitab 15 software. It was created a factorial design f (q) = 2n , being n=3 as it represents the experiment variables quantity chosen. The experiments design is shown on the Table 1. The variables were designated as follow:
Table 1
Iron Ore +1 (iron ore weighting 8 grams); -1 (no material)
Lubrication +1 (with water weighting 8 grams); -1 (no water)
The roughness parameter Ra before and after the experiments have shown low significant variation based on the average and standard deviation analysis. When submitted to the T-student average evaluation (Figure 2), it is not evidenced that both averages before and after the experiments are different: the P-Value of 0,707 represents the argument. Moreover, further on towards the tendency and normality charts at the Figure 3, it is verified that this behavior is considered statistically normal. Even though, according to the literature, the material type is what determines the amount of friction, not the surface finishing, (Hyun S Pei L., 2004). The total blade experiments travel on the rubber-belt is 1,60 meter.
Figure 2: T-Student roughness average
Figure 3: Tendency chart and normality chart of roughness values)
3.2. Design of Experiments (DOE)
The results shown on Table 2 and Figures 4 and 5 revealed a correlation between the factors and its individual contributions, named Significance, on the friction coefficient. As a reference value, the expected kinetic friction coefficient k in the literature is 1,16 (rubber-rubber) (Zhang, S.W, 2008). The weight needed to take the cart off the Inertia in its free state, no load, was discounted from the weights added to move the cart.
Table 2: Friction coefficients
The higher value is reached with the Experiment 4 where the Iron Ore is present with Lubricating water with the same weight of Iron, and with the higher pressure of 0,10MPa (Table 1). Meanwhile the second ranked friction coefficient is the Experiments 2, which opposes the n. 4 in terms of iron, lubrication and pressure. The lower friction coefficient is found in the Experiment 5 where there is no iron ore, but with lubrication (water) and under the higher pressure (0,10MPa). Roughly speaking, it would be expected lower friction coefficient without the presence of iron ore, with lubrication (water) and low pressure, as in the Experiment 6, rather than the Experiment 5 which is similar of it, but with the higher pressure. Despite of these, the difference is not considered too substantial. Visually, the ore iron presence looks like as helping the blade to slide on the rubber due to its cubic shape. It has shown that the higher pressure applied made the friction coefficient become lower due to the ore iron rolling. The significance of each variable on the experiments is shown in Figure 4 below. It is seen that the variable most significant in terms of friction is the ore iron followed by the lubricant and at last the applied pressure. The combination of ore iron with lubricating water is the key to produce the friction coefficient change, visually there is a combination between them which creates a mud (paste) and turn difficult the blade movement, increasing the friction coefficient (Figures 4 and 5). When comparing the Experiments 2 with 6 (Experiments with low pressure and no iron), regarding the lubrication, it is observed that the water drops the friction coefficient down, the same happens with the Experiments 7 and 5 but now under higher pressure.
Figure 4: Experiments variables significance based on friction coefficient
Figure 5: Experiments variable interactions based on the friction coefficient
Calculated for each factor and level, the main effect of the ith level of a factor in a balanced design is estimated by the ith factor level mean minus the overall mean. The main effect is:
yi y (1)
For the unbalanced case, Minitab uses the regression model to estimate the effects. For instance, in the two-factor mixed model, the main effect of the fixed factor A at the ith level is a i, which is estimated by the coefficient for a i, obtained after fitting the model, see Figure 4.
An interaction is present when the response at a factor level depends upon the level of other factors. If the factors effects are not addictive, they are interactive. For example, for a balanced two-way ANOVA model, the formula for interaction effect, ij , of the ith level of factor A and the jth level of factor B is:
ij= yij. - (yi.. - y.j. - y...) (2)
where yij. is the average of the observations at the ith level of factor A and the jth level of factor B, yi.. = average of observations at the ith factor level, y.j. is the average of observations at the jth factor level, and y... is the average of all observations.
For the unbalanced designs, Minitab 15 uses the regression model to estimate the effects.
For the interaction plot, the x-axis contains tick marks for each level of factor A and the y-axis contains the response. On the graph, Minitab plots a line for each level of factor B. Each endpoint of the line is equal to the mean response when factor B is set at the specified level and factor A is set at the specified level.
The Figure 5 reveals the interaction between the variables from the experiments. Comparing the Ore Iron with the lubricating we can argue that the dashed line which represents the Iron moves up when the water is present and moves down when absent. The close right square chart compares the interaction between the Ore iron presence with Pressure. When Iron Ore is present and with high pressure applied the friction line (dashed line) moves up while the opposite movement is found in contrary condition. The beneath square compares the Pressure with the Lubrication, when the pressure is higher and with water the friction coefficient moves up and low values are found when the system is dry and the pressure is higher. Those comparisons between each two variables are complemented by the previous Figure 4.
4.Conclusions
The DOE experimental method reveals some interesting correlations among the variables ore iron, lubricant and pressure on a blade against rubber-belt frictional system. Based on the experiments, some observations and conclusions are stated as follow.
-There is an interaction effect among the variables of the experiments
-the friction coefficient changed in some categories consistently with the added variables
-statistically the most significant and ranked variables on the system are ;
oOre Iron Presence;
oLubrication and ;
oPressure.
-The division in categories shown in the Table 2 reveals that the opposite tests (Experiments 4 and 2) regarding the variables settings presented a pretty close friction coefficient and highest ones.
-The category 3 (experiments #1, #3 and #5) had one variables changed each time and there was no significant friction coefficient change. It reinforces the interaction between the variables ore iron and the lubricant.
-The friction changed reached 100% with the parameters added on the system (Exp. 4 and 2).
-The Force applied, Normal Force to the rubber-belt surface, by itself was not able to change significantly the friction coefficient, and maybe it can be considered as having into the system error, not evaluated.
-The knowledge of those variables contribution to the friction system is of such importance on the scrapers used to clean up belt conveyors and other application where metallic materials and lubrication are present.
-The lubrication when combined with the Ore Iron generates nevertheless a paste (mud) which increases the friction coefficient, which difficult the movement.
-The highest wearing is expected to happen when the friction coefficient is higher and which based on the experiments dependable of Iron and Lubrication (water) and with an eligible pressure/force contribution. As described by (Zhang, S.W. 1998), the frictional coefficient is independent of applied load over a wide range.
-The Lubrication has dropped the friction coefficient down in the system without Iron and to both pressure applied.
-The increase on the polycarbonate blade pressure on the rubber-belt does not increase the friction coefficient by itself what means that the wearing is not pressure/force sensitive but Iron and Lubrication ones.
-During summer time, based on the experiments, the humidity increase contributes detrimentally to the scrapers performance.
5.Acknowledgements
The acknowledgements are addressed to CNPq (National Council for Scientific and Technological Development – Brazil), Mr. Wanderlei Sançon, Designer, Mr. Waldemar Cont Production Supervisor, Jos Martinez and Edilson Pereira from the Metso Technology Process Laboratory and Mr. Rubens Costa, Vice President, South America and Engineering Director Operations at Metso Brazil.
6.References
Zhang, S.W., (1998), State-of-the-art of polymer Tribology. Tribology International, vol. 31, 1-3, pp. 49-60.
Myshkin N.K. Petrokovets, M.I, Kovalev, (2005), A.V. Tribology of polymers: Adhesion, friction, wear and mass-transfer. Tribology International, Elsevier, 38, 910-921.
Bush A. W. Gibson R.D. and Thomas T R, (1975), Wear 35-87.
Yang C., Tartaglino U and Persson B. N.J., (2006) EUr. Phys.J.E. 19-47.
Grenwood J. A. and Willianson J B P, (1996), Proc. R. Soc. A 295, 300.
Hyun S Pei. L., Molinari J F and Robins M O, (2004) Phys. Rev. E 70 026117.
Lee, J.H., Xu, G.H., Liang, H. Experimental and numerical analysis of friction and wear behavior of polycarbonate. (2001) Wear 251, 1541-1556.
Campana C. and Müser M. H, (2007), Eur. Phys. Lett. 77 38005.
Shooter K., Thomas R.H. Frictional properties of some plastics. Research (1952); 2; 533-9.
Mikhalovich, G., Bartenev V., et al. Friction and wear of polymers, (1981), Elsevier.
Minitab 15, user manual, (2009).
Fig. 1: Device to carry out the friction experiment
Fig. 2: T-Student roughness average
Fig. 3: Tendency chart and normality chart of roughness values
Fig. 4: Experiments variables significance based on friction coefficient
Fig. 5: Experiments variable interactions based on the friction coefficient
Experimental Kinetic Friction Coefficient
Experimental kinetic friction coefficient (k) determination in the interface polycarbonate blade and flat rubber-belt when interacting with Ore, Lubricant and Pressure
Jairo A Martins
Department of Bucking Machines Handling and Conveyor Components
Engineeering, Metso Minerals, Av. Independncia, 2500 Iporanga, ZIP
18087-101 Sorocaba SP Brazil.
E-mail: jairo.martins@metso.com
*Corresponding author
Istvn Kövesdy
Department of Bucking Machines Handling and Conveyor Components
Engineeering, Metso Minerals, Av. Independncia, 2500 Iporanga, ZIP
18087-101 Sorocaba SP Brazil.
E-mail: istvan.kovesdy@metso.com
Jos D. Bressan
Center of Technological Sciences - CCT – Santa Catarina State University UDESC, Campus Prof. Avelino Marcante s/n - Bom Retiro - ZIP 89223-100,
Joinville-SC- Brazil
E-mail: dem2jdb@joinville.udesc.br
Abstract:
It is well known that even the most advanced Finite Element Analysis (FEA) version is not able to predict friction coefficients among materials. This paper presents an alternative and experimental method based on Design of Experiments (DOE) to determine the friction coefficient between polycarbonate blades against rubber-belt. Those experiments correlate Ore, Lubricating and Pressure in order to determine the friction coefficients in the interface between these materials with polycarbonate and rubber belt. The experiments results are of extreme importance for polycarbonate application on belt conveyor scraper and other applications involving polymeric surfaces under friction.
Keywords:
Friction Coefficient; Polycarbonate; Ore; Rubber Belt; Design of Experiments (DOE)
Reference:
J A Martins; Istvn Kövesdy and J.D.Bressan (2010). “Experimental kinetic friction coefficient (uk) determination in the interface polycarbonate plate and flat rubber-belt when interacting with Ore, Lubricant and Pressure, Int. J. Mining and Minerals Engineering, Vol 2, No 2, pp. 159-168.
(Published with the kind permission of the Int. J. Mining and Minerals Engineering)
1.Introduction
Polymers are becoming more widely employed as bearing and slider materials due to the availability and design flexibility of these materials. As polymers improve its properties and as designers learn the advantages of polymers application, public acceptance also increases. Friction is a very common phenomenon in daily life and in industry, which is governed by the chemical and physical phenomenon occurring in a thin surface layer of solid bodies in the moving contact (Zhang, S.W 1998). With the application of polymers for tribological purposes, increasing and extending to even more new areas, increasing research has been carried out on the Tribology of polymers and its application in various industries. A widespread interest in plastics has grown since the middle 20th century due to the features of their structure, design flexibility, specific mechanical behavior, and considerable possibility to change the polymer properties (Myshkin N.K. et al., 2005). Most studies of contact mechanics have focused on small loads where the contact depends linearly on the load (Bush A W, 1975 and Campana C. and Müser, 2007). Many polymeric materials have an excellent strength-weight ratio and often are easy to manufacture; some polymers also possess excellent tribological properties (Yang C., et. al,2006 and Shooter K., Thomas R.H., 1952). Polycarbonate is a widely used thermoplastic polymer for its high hardness and toughness either as a single phase material or as the matrix of a composite material (Grenwood J. A. and Willianson J. B. P., 1996). There have been fewer theoretical studies on the friction and wear behavior of polymers in general, and polycarbonate in particular (Hyun S. Pei. L, 2004). Glassy polymers are usually different from metals in many aspects, such as a larger elastic zone, elastic modulus dependent on temperature, pressure and strain rate sensibility, strain-hardening due to orientation of polymer chain (orientation hardening), strain softening and a more viscous behavior (Lee, J.H., Xu, G.H., et al., 2001). Due to the difficulty to test and determine friction coefficient in the field/application and neither by Finite Element Analysis these experiments were performed. This paper aims to replicate the scrapers concept used in conveyors but under a laboratory environmental and toward mathematical tool, Design of Experiments (DOE) (Minitab 15- Manual, 2009). This tool was used to plan, analyze and afterwards to determine the friction coefficient as well as the significance of each parameter [iron ore, lubrication (water) and pressure] on its value, occurred in the interface between polycarbonate and rubber-belt. This knowledge is of deep relevance on the design and operation of scrapers; polycarbonate made, and belt conveyors designs.
2.Materials and Method
2.1. Rubber-belt roughness
The measurement of roughness by the Ra parameter was done in order to verify the possible rubber-belt wearing variation before and after the experiments. The roughness measurements were performed on the rubber-belt surface in 9 points in the longitudinal as well as in the transversal direction.
2.2. Special device
In order to carry out the experiments, a special device was developed, according to attached Figure 1 below.
The Load 1 is added on the upper plate to press the polycarbonate blade against the rubber belt and the load 2 is hanging to make the rubber-belt to start the movement. The table where the rubber-belt is fixed is supported by ball bearings and pulled forward by a steel cable which turns the pulley (bearing) and further on engaged in the hanged Load 1. The rubber-belt travel is limited on 200mm. The test speed depends on the Load 2 but the weight value was selected for sliding speeds within the range of 0,01-1,0 cm/s, in which, according to Myshkin N.K, et al. 2005, the friction coefficient is speed-independent.
Figure 1: Device to carry out the friction experiment
2.3. Experimental parameters and DOE
The chosen parameters for the experiments were: Cubic Ore Iron size ranging from 150m to 300m, the Pressure applied by the superior weight (Load 1) was 0,04MPa and 0,10MPa and the Lubricant (water). The study of these variable combinations was done by running a design of experiment DOE method in the Minitab 15 software. It was created a factorial design f (q) = 2n , being n=3 as it represents the experiment variables quantity chosen. The experiments design is shown on the Table 1. The variables were designated as follow:
Table 1
Iron Ore +1 (iron ore weighting 8 grams); -1 (no material)
Lubrication +1 (with water weighting 8 grams); -1 (no water)
Pressure +1 (P = 0,10 MPa / F = 122 N); -1 (P = 0,04 MPa / F = 49 N)
3.Experimental Results
3.1. Roughness
The roughness parameter Ra before and after the experiments have shown low significant variation based on the average and standard deviation analysis. When submitted to the T-student average evaluation (Figure 2), it is not evidenced that both averages before and after the experiments are different: the P-Value of 0,707 represents the argument. Moreover, further on towards the tendency and normality charts at the Figure 3, it is verified that this behavior is considered statistically normal. Even though, according to the literature, the material type is what determines the amount of friction, not the surface finishing, (Hyun S Pei L., 2004). The total blade experiments travel on the rubber-belt is 1,60 meter.
Figure 2: T-Student roughness average
Figure 3: Tendency chart and normality chart of roughness values)
3.2. Design of Experiments (DOE)
The results shown on Table 2 and Figures 4 and 5 revealed a correlation between the factors and its individual contributions, named Significance, on the friction coefficient. As a reference value, the expected kinetic friction coefficient k in the literature is 1,16 (rubber-rubber) (Zhang, S.W, 2008). The weight needed to take the cart off the Inertia in its free state, no load, was discounted from the weights added to move the cart.
Table 2: Friction coefficients
The higher value is reached with the Experiment 4 where the Iron Ore is present with Lubricating water with the same weight of Iron, and with the higher pressure of 0,10MPa (Table 1). Meanwhile the second ranked friction coefficient is the Experiments 2, which opposes the n. 4 in terms of iron, lubrication and pressure. The lower friction coefficient is found in the Experiment 5 where there is no iron ore, but with lubrication (water) and under the higher pressure (0,10MPa). Roughly speaking, it would be expected lower friction coefficient without the presence of iron ore, with lubrication (water) and low pressure, as in the Experiment 6, rather than the Experiment 5 which is similar of it, but with the higher pressure. Despite of these, the difference is not considered too substantial. Visually, the ore iron presence looks like as helping the blade to slide on the rubber due to its cubic shape. It has shown that the higher pressure applied made the friction coefficient become lower due to the ore iron rolling. The significance of each variable on the experiments is shown in Figure 4 below. It is seen that the variable most significant in terms of friction is the ore iron followed by the lubricant and at last the applied pressure. The combination of ore iron with lubricating water is the key to produce the friction coefficient change, visually there is a combination between them which creates a mud (paste) and turn difficult the blade movement, increasing the friction coefficient (Figures 4 and 5). When comparing the Experiments 2 with 6 (Experiments with low pressure and no iron), regarding the lubrication, it is observed that the water drops the friction coefficient down, the same happens with the Experiments 7 and 5 but now under higher pressure.
Figure 4: Experiments variables significance based on friction coefficient
Figure 5: Experiments variable interactions based on the friction coefficient
yi y (1)
Calculated for each factor and level, the main effect of the ith level of a factor in a balanced design is estimated by the ith factor level mean minus the overall mean. The main effect is:
For the unbalanced case, Minitab uses the regression model to estimate the effects. For instance, in the two-factor mixed model, the main effect of the fixed factor A at the ith level is a i, which is estimated by the coefficient for a i, obtained after fitting the model, see Figure 4.
An interaction is present when the response at a factor level depends upon the level of other factors. If the factors effects are not addictive, they are interactive. For example, for a balanced two-way ANOVA model, the formula for interaction effect, ij , of the ith level of factor A and the jth level of factor B is:
where yij. is the average of the observations at the ith level of factor A and the jth level of factor B, yi.. = average of observations at the ith factor level, y.j. is the average of observations at the jth factor level, and y... is the average of all observations.
For the unbalanced designs, Minitab 15 uses the regression model to estimate the effects.
For the interaction plot, the x-axis contains tick marks for each level of factor A and the y-axis contains the response. On the graph, Minitab plots a line for each level of factor B. Each endpoint of the line is equal to the mean response when factor B is set at the specified level and factor A is set at the specified level.
The Figure 5 reveals the interaction between the variables from the experiments. Comparing the Ore Iron with the lubricating we can argue that the dashed line which represents the Iron moves up when the water is present and moves down when absent. The close right square chart compares the interaction between the Ore iron presence with Pressure. When Iron Ore is present and with high pressure applied the friction line (dashed line) moves up while the opposite movement is found in contrary condition. The beneath square compares the Pressure with the Lubrication, when the pressure is higher and with water the friction coefficient moves up and low values are found when the system is dry and the pressure is higher. Those comparisons between each two variables are complemented by the previous Figure 4.
4.Conclusions
The DOE experimental method reveals some interesting correlations among the variables ore iron, lubricant and pressure on a blade against rubber-belt frictional system. Based on the experiments, some observations and conclusions are stated as follow.
-There is an interaction effect among the variables of the experiments
-the friction coefficient changed in some categories consistently with the added variables
-statistically the most significant and ranked variables on the system are ;
oOre Iron Presence;
oLubrication and ;
oPressure.
-The division in categories shown in the Table 2 reveals that the opposite tests (Experiments 4 and 2) regarding the variables settings presented a pretty close friction coefficient and highest ones.
-The category 3 (experiments #1, #3 and #5) had one variables changed each time and there was no significant friction coefficient change. It reinforces the interaction between the variables ore iron and the lubricant.
-The friction changed reached 100% with the parameters added on the system (Exp. 4 and 2).
-The Force applied, Normal Force to the rubber-belt surface, by itself was not able to change significantly the friction coefficient, and maybe it can be considered as having into the system error, not evaluated.
-The knowledge of those variables contribution to the friction system is of such importance on the scrapers used to clean up belt conveyors and other application where metallic materials and lubrication are present.
-The lubrication when combined with the Ore Iron generates nevertheless a paste (mud) which increases the friction coefficient, which difficult the movement.
-The highest wearing is expected to happen when the friction coefficient is higher and which based on the experiments dependable of Iron and Lubrication (water) and with an eligible pressure/force contribution. As described by (Zhang, S.W. 1998), the frictional coefficient is independent of applied load over a wide range.
-The Lubrication has dropped the friction coefficient down in the system without Iron and to both pressure applied.
-The increase on the polycarbonate blade pressure on the rubber-belt does not increase the friction coefficient by itself what means that the wearing is not pressure/force sensitive but Iron and Lubrication ones.
-During summer time, based on the experiments, the humidity increase contributes detrimentally to the scrapers performance.
5.Acknowledgements
The acknowledgements are addressed to CNPq (National Council for Scientific and Technological Development – Brazil), Mr. Wanderlei Sançon, Designer, Mr. Waldemar Cont Production Supervisor, Jos Martinez and Edilson Pereira from the Metso Technology Process Laboratory and Mr. Rubens Costa, Vice President, South America and Engineering Director Operations at Metso Brazil.
6.References
Zhang, S.W., (1998), State-of-the-art of polymer Tribology. Tribology International, vol. 31, 1-3, pp. 49-60.
Myshkin N.K. Petrokovets, M.I, Kovalev, (2005), A.V. Tribology of polymers: Adhesion, friction, wear and mass-transfer. Tribology International, Elsevier, 38, 910-921.
Bush A. W. Gibson R.D. and Thomas T R, (1975), Wear 35-87.
Yang C., Tartaglino U and Persson B. N.J., (2006) EUr. Phys.J.E. 19-47.
Grenwood J. A. and Willianson J B P, (1996), Proc. R. Soc. A 295, 300.
Hyun S Pei. L., Molinari J F and Robins M O, (2004) Phys. Rev. E 70 026117.
Lee, J.H., Xu, G.H., Liang, H. Experimental and numerical analysis of friction and wear behavior of polycarbonate. (2001) Wear 251, 1541-1556.
Campana C. and Müser M. H, (2007), Eur. Phys. Lett. 77 38005.
Shooter K., Thomas R.H. Frictional properties of some plastics. Research (1952); 2; 533-9.
Mikhalovich, G., Bartenev V., et al. Friction and wear of polymers, (1981), Elsevier.
Minitab 15, user manual, (2009).
Fig. 1: Device to carry out the friction experiment
Fig. 2: T-Student roughness average
Fig. 3: Tendency chart and normality chart of roughness values
Fig. 4: Experiments variables significance based on friction coefficient
Fig. 5: Experiments variable interactions based on the friction coefficient
Table 1: Design of Experiments
Table 2: Friction coefficients
Attachments
metso_logo-1-1 (PNG)
■