Virtual Manufacturing Empowers Digital Product Development Case - PowerPoint PPT Presentation

Virtual Manufacturing Empowers Digital Product Development Case Study E-Coat Simulation Klaus Wechsler

Abstract Recent progress in simulation methods for the manufacturing industry has reduced the need for expensive test hardware which could be gratis used in manufacturing. Using manufacturing simulation tools starting at the design stage helps to optimize product development and corresponding manufacturing systems. Whenever there is a need for an early design input in order to ensure quality and manufacturing costs - virtual manufacturing methods will have a profitable chance. For the case study E-Coat simulation STAR-CCM provides an improved workflow from CAD-data and meshing to E-coat deposition as well as modeling of fill and drain behaviors in vehicle paint shops. Simulation results provide the design engineer with answers to questions such as ´is the E-Coat providing corrosion protection in all the cavities´? or ´is there a corrosion risk based on air bubbles or paint ponds´? The combination of an implemented fast algorithm with the chance of describing customer developed material properties by Field Functions allows best fit to complex chemical material behavior . Customized material development is kept inside customers. Based on process knowledge we are also providing multiple simulation support. An overview of future manufacturing simulation topics will be given. 2

Virtual Manufacturing is a Consequence of Hardware Reduced Digital Product Development Doing it in a physical way takes too long. Whenever there is a need for an early design input in order to ensure quality and manufacturing costs: Virtual Manufacturing methods will have a profitable chance. 3

The E-Coat Deposition Process Provides Corrosion Protection in all Cavities 4

Overview of the E-Coat Simulation Steps Data Freeze of BIW-Meshing Dipping in: E-Coat Dip-Tank Dipping Out Digital Product ( for ex. Wrapper, ( Air Bubbles , ( Puddles, Drainage - ( Paint Thickness on Development Boolean Unite) Pressure Distribution) surfaces and cavities) Time, Pressure Distr.) Suggestions for Design Optimization Goal: Corrosion protective E-Coat thickness, minimized air bubbles and puddles in all parts and cavities….) 5

E-Coat Modeling: Deposition of charge carrying paint polymers. Increasing resistance gives a chance for deposition inside cavities E-Coat Paint material : ..solids (pigments, resin/binder,.), solvent (de-ionized water) and co-solvents Electrolyte (glycol ether..) Conductivity is mainly based on the BIW=Cathode resin fraction but sensitive to carryover of conductive pre treatment materials from previous dipping steps .. Electro Static Paint 6

E-Coat Modeling: ..a small Chemical Plant.. Anode - Anode : Anolyte Circulation with influence on pH and film re-dissolution . - Dip Tank : Mixture of old (aged) and new material as well as recirculation from Rinse Tank. Needs permanent agitation and precise temperature control . - Pretreatment: Carryovers affect conductivity. 7

Equations of the STAR-CCM+ Standard Electro-Deposition Model ℎ dh c   dR dh Paint layer thickness in m      PL P PL PL j j r n min P dt dt dt ρ Paint layer resistance in Ω m 2 P   Specific electric current in A/m 2     j if q q j n if q q         0 j with q j n dt  n 0 j with q j dt  min 0  0 if q q  min n 0  0 if q q 0 Input Parameters of Standard Range of Values form Deposition Model calibration measurements 2-4· 10 -5 kg/As c P : Coulomb efficiency ρ P : Paint layer density 1200-1800 kg/m³ 2-5· 10 6 Ω m r P : Paint layer resistivity q 0 : Minimum Accumulated ´Activation Charge ´which 300-400 As/m² is necessary to start deposition in standard material) There is no deposition as long as q<q 0 σ liquid : Bath (Electrolyte) conductivity 0.14-0.22 S/m 8

Example of Enhanced (Customized) Electro-Deposition Model Using Field Functions for Detailed Calibration Measurements ℎ dh c   dR dh Paint layer thickness in m      PL P PL PL j j r n min P dt dt dt ρ Paint layer resistance in Ω m 2 P  Specific electric current in A/m 2  2 2 j if J J    2 2  n 0 j with J j dt Electric Potential at top of paint layer  min n ( ) 2 2  0 if J J in V 0 Cp(t) = f( ( ) , Input Parameters of Enhanced Deposition (t)) Model: - Variable Coulomb Efficiency C p (f b , C2 u , C1 u , C0 u = based on detailed - Deposition Starts if J 2 > J 0 2 calibration measurements) c P : Coulomb efficiency (1-exp(- ))*(f b *exp(- /h 0 ))+(-C2 U * U²+C1 U * U+C0 U ) ρ P : Paint layer density 1200-1800 kg/m³ 2-5· 10 6 Ω r P : Paint layer resistivity m 2 : Minimum Accumulated ´Activation Work ` (which is J 0 A 2 s/m 4 necessary to start deposition. There is no deposition as long as J 2 < J 0 2 σ liquid : Bath (Electrolyte) conductivity 0.14-0.22 S/m 9

Calibration of E-Coat Simulation Parameters: (1) Using Existing Real Parts ( if CAD Data are Available) Outside: Inner: x = 25 µm x = 18 µm Hidden: x = 8 µm Where can these data be found: E-Coat Thickness (µm) x - Sometimes paint shop regularly opens x = measurement parts for quality assurance x - Durability and other testing departments might have opened parts x Calibration: - Mesh real part and tank and adjust the parameters of the deposition model until ´best parameter fit´ is reached. Provides a fast pragmatic calibration with - Use conductivity and paint layer density focus to final (corrosion relevant) thickness from direct measurement/supplier. 10

Calibration of E-Coat Simulation Parameters: (2) Using Calibration Tubes Fixed to an Existing Part x = measurement Preparation: - Tubes are fixed to a part being coated - Tubes are opened and measured inside Calibration: - CAD model of tubes should be added to CAD model of part being simulated at a similar position. - adjust the parameters of the deposition Provides a pragmatic calibration with model until ´best fit´ of tubes is reached. standardized test geometry 11

Calibration of E-Coat Simulation Parameters: (3) Lab Measurements with Test Box and Plain sheets Box is closed on bottom and side. Top is above electrolyte level

Calibration of E-Coat Simulation Parameters: (3) Lab Measurements with Test Box (Medium Throw Power) 100V 200V Measurement Measurement

Calibration of E-Coat Simulation Parameters (3) Lab Measurements with Test Box (Good Throw Power) 100V 200V Measurement Measurement

Application of STAR-CCM+ E-Coat Simulation: Thickness Building over Time 15

Application of STAR-CCM+ E-Coat Simulation: Visualization of Thickness in Cavities Video Page 16

Application of STAR-CCM+ E-coat Simulation: Example of Good Throw Power 17

Application of STAR-CCM+ E-coat Simulation: Example of Medium Throw Power 18

Application of STAR-CCM+ E-coat Simulation: Example of Poor Throw Power 19

Application of STAR-CCM+ E-coat Simulation: Comparison of different Throw Power Capabilities Good Poor Medium 20

Application of STAR-CCM+ E-coat Simulation: Comparison of different Throw Power Capabilities Geometrical Variation Good will be necessary Poor Medium 21

Application of STAR-CCM+ E-coat Simulation: Evaluation of Gemetrical Variation (for better Corrosion Protection) Holes have influence on E-Coat thickness Bigger Diameter or more holes improve corrosion protection

Simulation of Dipping in: (1sec=1h on 32 CPU, 8 Cores/CPU) Remaining Air Bubbles Avoid E-Coat Film Building (Red = Trapped Air Bubbles) Video 23

Simulation of Dipping in (1sec=1h on 32 CPU, 8 Cores/CPU) Remaining Air Bubbles Avoid E-Coat Film Building 24

Simulation of Dipping in: (1sec=1h on 32 CPU, 8 Cores/CPU) Visualization of Trapped Air Video 25

Simulation of Dipping in (1sec=1h on 32 CPU, 8 Cores/CPU) Remaining Air Bubbles Avoid E-Coat Film Building 26

Simulation of Dipping in: Positioning of additional Bleeding Holes Final Position in E-coating should be without air bubbles. Simulation gives information for positioning of bleeding holes. 27

Simulation of Dipping in: Quick optimization check by adding holes and continuing simulation (Red = Trapped Air Bubbles) Holes added at t = 25s 28

Simulation of Dipping out: ( Remaining Ponds Contaminate Next Dipping Process Step) (Blue = Trapped Dipping Liquid) Video 29

Simulation of Dipping out: (1sec=1h on 32 CPU, 8 Cores/CPU) ( Remaining Ponds Contaminate Next Dipping Process Step) 30

Simulation of Dipping out: (1sec=1h on 32 CPU, 8 Cores/CPU) Calculation of Drainage Time 31 Video

Simulation of Dipping out: Quick optimization check by adding holes and continuing simulation: (Blue = Residual ponds) 32

Simulation of Dipping out: Quick optimization check by adding holes and continuing simulation: (Blue = Residual ponds) Holes added at t = 20s 33

Details of Dipping out Simulation: Remaining Ponds Contaminate next Dipping Process Step 34