A preliminary examination of blades wear in a planetary concrete mixer M.Cristina Valigi1, Ilaria Gasperini2. 1 Department of Industrial Engineering (DIIN), University of Perugia, Italy E-mail:email@example.com 2 SICOMA, Società Italiana Costruzione Macchine, Ponte Valleceppi (PG), Italy E-mail: firstname.lastname@example.org Keywords: Wear, concrete mixer, blades. SUMMARY. Performance optimization of a mixer is an issue of great significance in many industrial technologies: the wear is one of the main phenomena to forecast and test during the design and working of mechanical components, because, if mal-designed, they could be breakage and the machine working compromised. In this paper the authors present the wear problem of the planetary vertical concrete mixer in which the mixer components more suffering wear are the blades. In order to study the wear problem some mixing process models existing in literature are described ,,. The factors influencing mainly blades wear are: kinematics, friction between slurry and blades, friction due to the leakage between vessel and blades and also an inadequate use of machine. In addiction the authors study the movement impressed to the mixer by the epicicloydal gear and the kinematics analysis of blades and arms. As consequence of the kinematics analysis a new shape of blade is proposed. 1 INTRODUCTION The wear effect is the progressive modification of surfaces morphology that brings, in a short time, the mechanical components to the final inefficiency. In the examined concrete mixer the blades result the components more suffering the wear. The mixing function does not appertain to the blades, but their function is unload the mixer, pushing the concrete to discharge opening. They are made so that they can be easily changed when the wear is proceeded and the unload function is almost inhibited. In order to extend the life of the blades, a new shape is proposed studying the mixing process and its characterization. The mixer power evolution with respect to operating time, describes the process as presented in literature .Particles were wetted and a power consumption increased. Thereafter, power consumption leveled off at a plateau and then further increased during the wet phase. In the final phase, power consumption drops as the mass becomes a suspension (pseudo-fluid). In the power absorption diagram, phases could be localized in which particular transformations occur. Many authors ,,,. have proposed models of mixture evolution correlated with the type of microstructure in various states stages (see Figure 1). In order to improve the performance of the mixer some authors  have proposed an analytical model for force estimation on arms, that is very useful for mechanical design. In this paper the main task is to analyze the wear of the blades and propose new shape with high wear resistance. Figure 1: Mixture evolution with type of microstructure in various states stages. 2 CONCRETE MIXER DESCRIPTION The concrete mixer analyzed in this paper is a batch mixer with a vertical axis and fixed tank . The working rotation is a planetary motion obtained by a gear unit with one or two stars and three mixing arms for each one (the Figure 2 shows a mixer with two motors, two stars and six arms). The mixing motion is obtained by the composition of gear rotation and arms revolution and its aim is to physically and chemically mix multiple components in order to make a homogeneous mixture. Figure.2 Planetary concrete mixer. Actually the mixer distributes all the constituents uniformly in the tank without favoring one or the other. The blades (one for each arm) can reach all points of surface tank in a short time as the Figure 3 shows. Figure 3:Trajectories of arms at different times. The arms have the function to mix but the blades have the function of unload, for this reason the efficiency of mixing is not compromise by the wear or breaking of blades. The wall and the bottom of tank are interchangeable because elements subject to wear. The mixing tank is constructed in extremely thick sheet steel mounted on a channel section frame arranged with several discharge openings. The entire tank is protected by a casing to prevent dust escaping and the mixing operation will be interrupted by a micro-switch if the door at the front is opened. One or two peripheral arms work as scrapers. The slurry level has to be about one third of thank deep. 3. MIXING PROCESS CARACTERIZATION AND MODEL Fresh concrete is a granular fluid: a complex mixture of sand, gravel, cement, water and eventual chemical admixtures. A typical mixing schedule of the analyzed mixer is the following : Time (s) Components/ 0-5 Idling 5-20 Aggregate 20-40 Cement 40-50 Water 50–70 Mixing period 70-80 Discharge period The mixer has to start empty before the loading operation, the loading of water has to occur in a short time in order to gain the production time. Therefore a fast loading of water means a short time in which the mixer is subject to the largest effort (the absorption power reaches the 120% of nominal value). The final mix time depend on the time of previous phases. Regarding the concrete mixing operation, we can pick out three principal phases : Phase 1: first mixing phase, aggregate and cement are inserted in the mixer without water, “dry – phase”, Phase 2: a proper quantity of water is added to the dry mix for final mixing, “wet phase”, Phase 3: final mixing of concrete before unloading for complete homogenization of the mix. Some authors have shown how the Brinch-Hansen formula (derived from Terzaghi theoretical model)  and the Bingham theory of fluids  can be used to describe the behavior of concrete respectively in the so called dry and final mixing phases. For the second phase there is not a consistent theoretical model for the materials because the water addition time is usually very short compare to the time of other phases. Figure 4 shows the power absorption of the motor during a typical mixing cycle. In the power absorption chart two stages are highlighted during mixing: - At the beginning of the mixing cycle, the power absorption has strong and rapid variations that results as a competition between excess liquid zones and unsatured zone. - After a given mixing time, the mixture has already formed as a fluid-like material and the evolution of power drops to become smooth. The moment when the power consumption decrease suddenly is the time of transition from a cohesive mixture to a ﬂuid-like material. After the transition time, the flow is governed by viscous effects . transition time Figure 4: Typical power absorption of motor in the time for the analyzed mixer. Some different causes of electrical power absorption during the mixing can be individuate. The energy is dissipated in mixture flow mainly by friction for the dry granules states or dry power, by cohesion when granules are wetted in surfaces and by viscous effect in the granular suspension state. Cazacliu et al. have modeled  the origin of power absorption, with the forces that counters blade progression with two components: frictional Sf and viscous Sv. S d L f 0 (1) The frictional force is proportional to the contact between the surface of blades and the tank, that is present in the bottom zone and lateral zone. Where τ0 is the frictional tangential action depending on the type of the concrete in mixing, d is the distance between the tank and the blade surfaces and L is the characteristic dimension of blade ( that is the length). The viscous component Sv is function of the concrete apparent viscosity that is, according to the Bingham formula, the following: * 0 (2) And where is the velocity gradient in the blade vicinity: * S s L n v (3) v where s is a viscous coefficient and n the normal to the blade surface. (4) The power consumed by a blade is: P S S v v f According to the models present in literature the concrete is considered a polifasic fluid with a behavior as a Bingham material. A Bingham material is a viscoplastic material that behaves as a rigid body at low stress but flows as a viscous fluid at high stress (Figure 5). Figure 5: Comparison between Bingham and Newtonian fluid. The shear yield stress τo depends on the friction and cohesion stress of slurry, on attraction of concrete particles, on friction stress and on ratio water/concrete and it influences the slump. The plastic viscosity μ depends on the quantity and on the type of idles. 3 WEAR BLADES The blades suffering wear more than others parts of mixer because are completely invested by the material in mixing. The part of blades that results more worn is the extreme one that works in the circumferential zone of tank, that part has a fundamental role during the unloading phase because push the concrete against the discharge openings. In Figure 6 there are two blades: a new one and a blade that is at the end of its life. The wear in the blade is more evident in the right zone (that is the extremity of the blades). This behavior is typical for all the blades. The material of mixer blades is cast-iron Ny-Hard with hardness of about 550-600 HB, with a high wear resistance. The composition of alloy is the following: 3.25%C, 1.59%Si, 1.44Mn, 0.075%S, 2.86%Ni, 1.54%Cr. Figure 6: A blade at the end of its life and a new one. 3.1 Wear mechanism In the interaction between concrete and blade, the concrete could be considered as a polyfasic fluid with solid particles of different dimension. Because the solid is together with the liquid component, two different damage mechanisms could be generated by the relative movement of the fresh concrete respect to the blade mechanic action and corrosive action. The two actions can be summarized synergistically in case that the fluent fresh concrete is moved at high speed. Indeed the mechanic action can damage the passivity film protecting the material causing corrosive action at high velocity in the zones more protect by film. The corrosive action can occur quickly until to the protection of oxide film forms again, and then restarts because the continues and repeated damage due to the fluid movement. The mechanic action becomes sufficient when the value of critic speed is exceeded. The mechanical wear is erosive-abrasive due to different particles that during the impact remove material and that depends on its shape and dimension, on impact angle and relative velocity respect to the blade. The wear also depends on the characteristic property of blade material: in particular the hardness of idles has a direct rule in the abrasive mechanisms: the mean hardness of idles is about 4÷5 Mohs that is in Brinell about 400 MPa. The cast-iron Ny-Hard hardness is about 550÷600 MPa that is greater than the idles one. Therefore the mechanical action is erosive rather than abrasive. In addiction the corrosive action can occur for the contemporary presence of watery phase (the slurry water) and oxygen. The pH of watery solution arises rapidly because the formation of calcium hydroxide during the hydration process of concrete. In particular for steel and also cast-iron (even if to in lesser degree) the corrosion depend on the rotational speed in the slurry. Actually the corrosive action, increase with speed greater than 3m/sec. The Figure 7 shows the erosive wear in the extreme zone of the blade with a group of impacts with low angle (in blue) that are visible because they leave trials on material and impacts with high angle that produce out and out holes (in red). Low Impact Angle High Impact Angle Figure 7 : Erosive wear with low and high angle of impact. 4 PROPOSED SOLUTIONS Because the wear regards prevalently the right extremity, at beginning, the blades were modified by SICOMA with a thickening of that side in order to increase the wear resistance. Figure 8: Blades without and with thickening. Figure 8 shows the comparison of blades with and without thickening. This modification was a result of the collaboration between customers and firm. In this way the goal was reached, but in order to improve further the performance of mixer in this paper a new shape of blade is proposed on the basis of theoretical models described. 4.1 Proposal of a new shape of blade The blades are assembled on the mixer with an angle of 22,5° respect to the arm (see Figure 9), in the past years experimental tests were carried on blades assembled with angle of 0°. The results showed blades more resistant respect to the wear but with the unload function compromised. On the basis of the mentioned results and on theoretical model showed, a new shape of blade is proposed (see Figure 10). Figure 9: Blade without thickening presently assembled on the mixer. Figure 10. New proposed blade. The new blade has a pleat: in the internal part it is like the current blade (with the same angle), but since the middle of the front line it has a bigger angle. 4.2 Cinematic analysis of the new blade In order to study the wear the authors have assumed that the wear occurs, mainly, during the Phase 3 (final mixing of concrete before unloading for complete homogenization of the mix) that is the more important phase of mix process in terms of power absorption. According to the mentioned models, in the Phase 3 the slurry can be modeled as Bingham fluid so that the wear, in the extreme part of blades, is due, mainly, to tangential actions of the fluid. Therefore the new blade with different angles are proposed and an cinematic analysis is carried on and compared to the actual one. The characteristic of the new blade is that they have a lower peripherical speed that means lower viscous force acting on the blades and lower wear. The origin of Cartesian axis is choice in the center of the mixer, the mixer analyzed has one star and the cinematic analysis is carried on for one blade. The blade is represented by the line of its front and in order to evaluate the speed along the blade, five point have been considered. In the following figures are showed results for a blade with one of the proposed new angle. Figure 11 shows trajectories of the actual and the proposed blade in 5 different points of the front line . (a) (b) Figure 11. Trajectories of the actual (a) and the proposed blade (b) in 5 different points of the front line. (a) (b) Figure 12. Speed of the actual (a) and the proposed blade(b) in 5 different points of the front line . Figure 12 shows the speed of actual and proposed blade in five different points of the front line and Figure 13 the speed along the frontline. The speed is lower in the extreme of the proposed blade compare to the actual blade; for this reason some prototype blades are made forecasting a better resistance wear because a lower viscous force. (a) (b) Figure 13. Speed of the actual (a) and the proposed blade (b) along the line defining the blade. 5 RESULTS AND CONCLUSIONS On the cinematic analysis some prototype blades are made and an experimental test campaign is began in order to evaluate resistance wear and unloading time. Until now, measures of wear and of unload times show encouraging results: the new blades have a higher wear resistance and unload time shorter respect to the actual ones. The result of experimental test will be useful for to set up a model of blades wear in order to choose the best angle for the pleat. References  B. Cazacliu, J. Legrandb., “Characterization of the granular-to-fluid state process during mixing by power evolution in a planetary concrete mixer,” Chemical Engineering Science 63,(2008), 4617-4630.  B. Cazacliu , N. Roquet., "Concrete mixing kinetics by means of power measurement ” Cement and Concrete Reserch 39, (2009) 182–194.  C.Braccesi,L.Landi "An analytical model for force estimation on arms of concrete mixers” Proceedings of the ASME 2009 International DETEC/CIE,2009, San Diego, California, USA.  H.P.Bier, H . Leuenbeger, , H. Sucker, “Determination of the uncritical quality of granulating liquid by power measurements on planetary mixers.” Pharmazeutische Industrie 41 (4),(1979) 375–380.  B. Cazacliu,"In-mixer measurement to describe the mixture kinetics during concrete mixing" Proceedings of sixty International on Mixing in Industrial Process Industries,I SMIP VI , 2008,Niagara on the Lake, Niagara Falls, Ontario,Canada. [6 ] B. Cazacliu,"In-mixer measurement for describing the mixture evolution during concrete mixing"Chemical Engineering Research and Design 86,(2008), 1423-1433  A.Goldszal, J. , Bousquet, , “Wet agglomeration of powders: from physics toward process optimization.” Powder Technology 117,(2001) 221–231.  M.C. Valigi , I.Gapserini, “Planetary vertical concrete mixers: Simulation and predicting useful life in steady states and in perturbed conditions” Simulation modeling Practice and Theory.15 (2007) 1211-1223.