New Material Design Leads to an Other Casting Quality Solidification, Stress, Mechanical Properties. Konrad Weiss Christoph Honsel Reinhard Vomhof

Transcription

1 GTT-Technologies' 12th Annual Workshop, Herzogenrath, Germany, June 16-18, 2010 New Material Design Leads to an Other Casting Quality Solidification, Stress, Mechanical Properties Konrad Weiss Christoph Honsel Reinhard Vomhof RWP GmbH Germany Roetgen Bundesstr

2 October 1984 Foundation of RWP Pioneer in Simulation of Casting Process 2

3 Topics Introduction Facts of RWP Increasing use of computer for the component development The tendency of callback in the automotive industry is growing Current Situation for the Development of Castings Using CAE-Tools Using WinCast Virtual component development- Possible results in the moment Improvement of the forecast of mechanical characteristics Micro FE Micro Tensile Test Mono Materials Tensile Test Synthetic Rebuilding of a Real Micro Structure Example Stub Axle Conclusion 3

4 Facts of RWP 1984 RWP was founded after basic developments at the RWTH Aachen done by Dr. Konrad Weiß High experienced software company developing casting simulation software WinCast and SIMTEC are our software products for simulation of casting and cast processes RWP has become a Global Company with branch offices in USA, China, Russia, Spain, Czech Republic, Korea, India etc. RWP is an expert in casting engineering and supplies you with technical seminars, engineering services and software developments RWP works as a partner of industry and universities for external development and research projects on national and international level Main industrial fields of RWP are design and construction of optimized components, casting of metal and plastics, foundry process simulation and the simulation of welding, laser welding and process optimization 4

5 Introduction Increasing use of computer for component development Possible application areas for software : CAD - construction Topology optimization Stability- and Structural analysis Life cycle calculation Production analysis mould fill simulation cast and solidification process heat treatment and machining Crash analysis Vibration- und Acoustic analysis Flow analysis Multiple component system analysis Electromagnetism analysis 5

6 Introduction The tendency of callback in the automotive industry is growing. Different reasons : more hightech with higher risk of failure shorten model cycles as well as shorten development times tight calculated development and production costs Possible way of solution : Toyota Callback of 1,8 Million cars due to clamped gas pedals Sometimes it itseems that thatthe thecar [Quelle: cardriver Bild.de ] will will be bemisused as as a test test driver. A. DEMMEL (Leader automotive technique ADAC) virtual Product development drastic reduction of development time saving on Prototypes efficient and faster introduction of serial production Saving of time, costs and material during the development process No. of callbacks during the last 10 years [Source: ADACmotorwelt 12 / 2004] 6

7 Current Situation for Development of Castings by using CAE Sp A99 Database 120,00 100,00 80,00 60,00 40,00 20,00 0,00 0 0,0005 0,001 0,0015 0,002 0,0025 A99 The used physical properties are based on standard tensile tests => Homogenous Material Stress- Elongation-Diagramm Design CAE Optimization Optimization Release 01/02/.. Tests Final product - High safety factor - High weight - High costs - Quality level not easy to increase Conclusion Due to the dependency of current material information no adequate forecast of the mechanical behaviour of the component 7

8 Current Situation for Development of Castings by using WinCast A99 Database Sp 120,00 100,00 80,00 60,00 40,00 20,00 0,00 0 0,0005 0,001 0,0015 0,002 0,0025 A99 Taking the material properties into account due to solidification time Stress- Elongation-Diagramm Carbon-Iron-Phase-Diagramm Al-Si- Phase-Diagramm Design CAE Release 01/02/.. Tests Final product - (Still) High safety factor - Weight can be reduced - High costs - Quality level can be increased Optimization Optimization Conclusion Still dependency of standard material information so the forecast of mechanical properties still are not adequate enough 8

9 Virtual Component developmentpossible results in the moment Virtual component development of a cylinder head Life cycle prediction Prototype introduction of series production CADCAD-Construction Topology optimization Structural optimization FEMFEM-StabilityStabilitycalculation Stress simulation Casting process Heat treatment Machining Determination of mechanical Properties Mould fillfill- and solidification simulation Basics 9 WinCast The full line of Finite Element Simulation R.Vomhof,

10 Virtual Component development- Possible results in the moment The CAD data are the starting point The macroscopic simulation is used for form filling and solidification The result of the macroscopic temperature calculation can be used for the prediction of macroscopic mechanical properties Residual and load stress can be superimposed in order to evaluate the limits for strength fatique The prototype building is the target 10

11 Virtual Component development- Possible results in the moment CAD Data CAD Data is available as solid models and can be used for simulation after transformation into the STL- Format CAD Data is generally already available during the various component development steps. Initial simulation results for the casting process chain can be achieved even with a low degree of detail of the component. 11

12 Virtual Component development- Possible results in the moment Form filling and solidification Calculations can only be done in the area of m. Calculation of form filling and solidification is achieved using differential equations (Navier-Stokes, Fourier) Physical data is necessary for the simulation. Micro-models are necessary in order to determine solidification phenomina. 12

13 Virtual Component development- Possible results in the moment Structure/Microstructure Initial calculation of structures and microstructures are carried out on various levels. The macroscopical statements were developed using empirical investigations. The results are very close to those of investigations on real. The various models are in development and need more effort in research and development. 13

14 Virtual Component development- Possible results in the moment Residual- and Load stresses Various physical/mechanical characteristic data are required for consideration of the local mechanical characteristics within the. The residual stresses are the result of the cooling sequence during casting and solidification. The load stresses result of the calculation of the various load situations. The calculation of these various data can be achieved via virtual material development. 14

15 Virtual Component development- Possible results in the moment Life cycle definition Life cycle definition is achieved by comparing the various load collectives with the local stress cycle diagram. The calculation of the local stress cycle diagrams is developed from the macroscopic temperature profile. Solidification conditions and composition must be considered when calculating the local stress cycle diagrams. 15

16 Virtual Component development- Possible results in the moment Prototype / Series production Prototyping will continue to be necessary in the future, however, the number of components will be greatly reduced due to the capabilities of virtual component testing. The effort put into producing components after the casting process chain are not only usable for the development of prototypes. Simulation of the production chain, and therefore the influence on the appropriate component quality areas, deliver results which are used for the design of components, tools and processes. 16

17 Improvement of the forecast of mechanical characteristics The consequence of different wall thicknesses and local conditions leads to different temperature gradients and solidifications times Micro structure is not homogenous Distribution of Mechanical properties are within a range What must be done? The demand to get more and more information about the material properties already in the design stage of a new component leads to the necessity to know more about the building mechanism of the basic matrix. Here we give one important step to get more detailed information on the mechanism of building different mechanical properties. 17

18 Improvement of the forecast of mechanical characteristics The classical approach for the evaluation of data ist timeconsuming and strongly dependent on test conditions. New alternatives have to be found in order to be able to describe the various structures and the local conditions within the components. Lets take a different route 250 [N mm ] Spannung σ R p0,2 = p0,2 (175 ± R m (236 ± 5) 5) MPa R p0,2 = (175 ±13) MPa R R m = (236 MPa m = (236 ±5) MPa ±5) MPa R p0,2 = (175 ±13) MPa E = E (74 = ± 3) GPa E (74 = (74 ±3) GPa ±3) GPa A = A (1,0 = ± A (1,0 = (1,0 ±0,1 ±0,1) 0,1) ) % 0 0 0,3 0,6 0,9 1,2 [% 1,5 ] 1,8 Dehnung ε We create various virtual basic structures and test their characteristics. The characteristics are validated with targeted experiments. Constrution of synthetic structures Comparison of real and synthetic structures Creation of a correlation between solidification conditions and synthetic structures. Creation of a database for the calculation of the expected mechanical properties. 18

19 Improvement of the forecast of mechanical characteristics Development of tools Tool 1 The coupling of micro- and macro-simulation with the consequences on the solidifiction sequences combines nucleation and growth as well as the release of latent heat with the expected structures. Tool 2 Micro-FE shows the possibilities and limits of the various materials and processes. This method can also be used to create missing physical and mechanical data. 19

20 Micro-FE With the cristal, eutectic, matrix calculation we can see what happens during solidification. The resulting mechanical properties can not be defined. Therefore we will have a look at the micro FE analysis to get more information about the relationship between microstructure and mechanical properties. Sp 250,00 200,00 150,00 100,00 50,00 Mg, A99, Si, Fe Database 0,00 0 0,001 0,002 0,003 0,004 0,005 0,5 mm Micro-FE Mg A99 Si Fe 0,1 mm 0,1 mm Basics 20

21 Micro Tensile Test The tests of the mechanical properties are done with a micro tensile test. 0,5 mm Fixed points for analysis 0,1 mm 0,1 mm tests 21

22 Mono Materials [N/mm 2 ] Mg, A99, Si, Fe Database 250,00 200,00 Sp 150,00 100,00 Mg A99 Si Fe 50,00 0,00 0 0,001 0,002 0,003 0,004 0,005 tests 22

23 Rebuilding of Real Micro Structure We are interested in the mechanical properties in specific areas of a real microstructure. We need to know how the influence of the silica in the Alpha phase will be( A), also we would like the influence of the porosities(b). B A Model building 23

24 Rebuilding of Real Micro Structure We build the finite Element mesh to have a description of the matrix. The red areas represent the pores. A With this method it is possible to see the influence of the shape of porosities. Model building 24

25 Rebuilding of Real Micro Structure Here we just bring in the silica distribution into the Alpha matrix. Different shapes and sizes can be built. B Model building 25

26 Rebuilding of Real Micro Structure The view through the testing part shows the distribution of the different disturbants. This is possible In all directions. Model building 26

27 Rebuilding of Real Micro Structure Tensile Test synthetic Alloy synt 6 with real distribution A99 Si 6 % Si Alloy synt 7 with real distribution A99 Si 6 % and porosities Alloy synt 8 with real distribution A99 Si 7 % Si across Porosity Tensile test alloy synt no 6-8 to compare real distribution of alloying elements 27

28 Rebuilding of Real Micro Structure Influence of real distribution of Si and porosities [N/mm 2 ] 250,00 Comp A99 vs A99 Si real Distribution Alloy synt 8 (magenta) 200,00 150,00 Alloy synt 6 (yellow) A99 Sp 100,00 A99 3D Si 7% A99 S6 % 50,00 A99 Data Base 0,00 0 0,0005 0,001 0,

29 Example Stub Axle 29

30 Example Stub Axle Mould filling animation 30 WinCast The full line of Finite Element Simulation R.Vomhof,

31 Example Stub Axle Temperature field animation 31 WinCast The full line of Finite Element Simulation R.Vomhof,

32 Example Stub Axle Defect building animation 32 WinCast The full line of Finite Element Simulation R.Vomhof,

33 Example Stub Axle Solidification time 33

34 Example Stub Axle Hardness distribution 34

35 Example Stub Axle Tensile strength distribution 35

36 Example Stub Axle Principle stress distribution with respect to classical mech. properties 36

37 Example Stub Axle Principle stress distribution with respect to synthetical mech. properties 37

38 Conclusion Virtual componetdevelopment with Micro FE FEM-Stability Stability- Prototype calculation introduction of Series production Determination of mechanical properties Life cycle prediction Stress simulation Matrix determination Micro FE CAD-Construction Construction Mould fill- and solidification simulation Defect calculation and graphit expansion Conclusion 38

39 Further work to do Determination of real matrix distribution Determination of real mech. properties Determination of matrix solidification time relationship Creation of a building set for different alloys Etc. 39

40 Thank You 40